{"build": {"built_at": "2026-09-09T21:18:47+00:00", "commit": "f871156937ea0b1a9bd817d3ec81b9509167b20f", "eligibility_as_of": "2026-09-09"}, "candidates": [{"benchmark_scores": {"arc_challenge": 64.4, "bbh": 60.2, "gpqa_diamond": 36.3, "gsm8k": 0.0, "hellaswag": 85.5, "ifeval": 31.2, "math_500": 15.4, "mmlu_abstract_algebra": 56.0, "mmlu_anatomy": 75.6, "mmlu_astronomy": 89.5, "mmlu_business_ethics": 82.0, "mmlu_clinical_knowledge": 82.6, "mmlu_college_biology": 86.1, "mmlu_college_chemistry": 52.0, "mmlu_college_computer_science": 72.0, "mmlu_college_mathematics": 54.0, "mmlu_college_medicine": 74.6, "mmlu_college_physics": 59.8, "mmlu_computer_security": 86.0, "mmlu_conceptual_physics": 84.7, "mmlu_econometrics": 63.2, "mmlu_electrical_engineering": 80.7, "mmlu_elementary_mathematics": 74.3, "mmlu_formal_logic": 62.7, "mmlu_global_facts": 54.0, "mmlu_high_school_biology": 89.7, "mmlu_high_school_chemistry": 70.9, "mmlu_high_school_computer_science": 84.0, "mmlu_high_school_european_history": 88.5, "mmlu_high_school_geography": 91.9, "mmlu_high_school_government_and_politics": 94.8, "mmlu_high_school_macroeconomics": 83.6, "mmlu_high_school_mathematics": 51.9, "mmlu_high_school_microeconomics": 88.7, "mmlu_high_school_physics": 54.3, "mmlu_high_school_psychology": 91.0, "mmlu_high_school_statistics": 71.3, "mmlu_high_school_us_history": 89.2, "mmlu_high_school_world_history": 89.0, "mmlu_human_aging": 82.5, "mmlu_human_sexuality": 88.5, "mmlu_international_law": 88.4, "mmlu_jurisprudence": 88.9, "mmlu_logical_fallacies": 86.5, "mmlu_machine_learning": 58.9, "mmlu_management": 86.4, "mmlu_marketing": 95.3, "mmlu_medical_genetics": 87.0, "mmlu_miscellaneous": 90.5, "mmlu_moral_disputes": 82.1, "mmlu_moral_scenarios": 67.2, "mmlu_nutrition": 83.7, "mmlu_philosophy": 80.4, "mmlu_prehistory": 86.1, "mmlu_pro": 47.1, "mmlu_professional_accounting": 64.2, "mmlu_professional_law": 58.5, "mmlu_professional_medicine": 82.7, "mmlu_professional_psychology": 81.7, "mmlu_public_relations": 75.5, "mmlu_security_studies": 81.6, "mmlu_sociology": 89.1, "mmlu_us_foreign_policy": 92.0, "mmlu_virology": 57.8, "mmlu_world_religions": 90.6, "musr": 44.0, "truthfulqa": 52.1, "winogrande": 84.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.6, "display_name": "Yi 1.5 34B 32K", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "01-ai/yi-1-5-34b-32k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 70.5, "bbh": 61.0, "gpqa_diamond": 33.8, "gsm8k": 71.6, "hellaswag": 86.0, "ifeval": 45.6, "math_500": 21.4, "mmlu_abstract_algebra": 58.0, "mmlu_anatomy": 80.0, "mmlu_astronomy": 87.5, "mmlu_business_ethics": 81.0, "mmlu_clinical_knowledge": 77.4, "mmlu_college_biology": 82.6, "mmlu_college_chemistry": 56.0, "mmlu_college_computer_science": 69.0, "mmlu_college_mathematics": 51.0, "mmlu_college_medicine": 75.1, "mmlu_college_physics": 52.0, "mmlu_computer_security": 82.0, "mmlu_conceptual_physics": 81.7, "mmlu_econometrics": 57.9, "mmlu_electrical_engineering": 74.5, "mmlu_elementary_mathematics": 75.4, "mmlu_formal_logic": 66.7, "mmlu_global_facts": 51.0, "mmlu_high_school_biology": 88.4, "mmlu_high_school_chemistry": 66.5, "mmlu_high_school_computer_science": 90.0, "mmlu_high_school_european_history": 86.1, "mmlu_high_school_geography": 88.4, "mmlu_high_school_government_and_politics": 95.9, "mmlu_high_school_macroeconomics": 83.8, "mmlu_high_school_mathematics": 55.6, "mmlu_high_school_microeconomics": 87.8, "mmlu_high_school_physics": 60.9, "mmlu_high_school_psychology": 89.0, "mmlu_high_school_statistics": 68.1, "mmlu_high_school_us_history": 90.7, "mmlu_high_school_world_history": 88.2, "mmlu_human_aging": 80.7, "mmlu_human_sexuality": 87.8, "mmlu_international_law": 87.6, "mmlu_jurisprudence": 83.3, "mmlu_logical_fallacies": 85.9, "mmlu_machine_learning": 63.4, "mmlu_management": 88.3, "mmlu_marketing": 92.3, "mmlu_medical_genetics": 84.0, "mmlu_miscellaneous": 89.9, "mmlu_moral_disputes": 81.2, "mmlu_moral_scenarios": 68.2, "mmlu_nutrition": 82.7, "mmlu_philosophy": 77.5, "mmlu_prehistory": 85.2, "mmlu_pro": 45.4, "mmlu_professional_accounting": 63.1, "mmlu_professional_law": 57.0, "mmlu_professional_medicine": 81.2, "mmlu_professional_psychology": 78.3, "mmlu_public_relations": 72.7, "mmlu_security_studies": 82.0, "mmlu_sociology": 88.1, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 62.7, "mmlu_world_religions": 86.5, "musr": 44.0, "truthfulqa": 62.2, "winogrande": 81.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 16384, "cost_input": 0.6, "display_name": "Yi 1.5 34B Chat 16K", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "01-ai/yi-1-5-34b-chat-16k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 70.5, "bbh": 48.2, "gpqa_diamond": 36.5, "gsm8k": 71.6, "hellaswag": 86.0, "ifeval": 77.3, "math_500": 58.5, "mmlu_abstract_algebra": 58.0, "mmlu_anatomy": 80.0, "mmlu_astronomy": 87.5, "mmlu_business_ethics": 81.0, "mmlu_clinical_knowledge": 77.4, "mmlu_college_biology": 82.6, "mmlu_college_chemistry": 56.0, "mmlu_college_computer_science": 69.0, "mmlu_college_mathematics": 51.0, "mmlu_college_medicine": 75.1, "mmlu_college_physics": 52.0, "mmlu_computer_security": 82.0, "mmlu_conceptual_physics": 81.7, "mmlu_econometrics": 57.9, "mmlu_electrical_engineering": 74.5, "mmlu_elementary_mathematics": 75.4, "mmlu_formal_logic": 66.7, "mmlu_global_facts": 51.0, "mmlu_high_school_biology": 88.4, "mmlu_high_school_chemistry": 66.5, "mmlu_high_school_computer_science": 90.0, "mmlu_high_school_european_history": 86.1, "mmlu_high_school_geography": 88.4, "mmlu_high_school_government_and_politics": 95.9, "mmlu_high_school_macroeconomics": 83.8, "mmlu_high_school_mathematics": 55.6, "mmlu_high_school_microeconomics": 87.8, "mmlu_high_school_physics": 60.9, "mmlu_high_school_psychology": 89.0, "mmlu_high_school_statistics": 68.1, "mmlu_high_school_us_history": 90.7, "mmlu_high_school_world_history": 88.2, "mmlu_human_aging": 80.7, "mmlu_human_sexuality": 87.8, "mmlu_international_law": 87.6, "mmlu_jurisprudence": 83.3, "mmlu_logical_fallacies": 85.9, "mmlu_machine_learning": 63.4, "mmlu_management": 88.3, "mmlu_marketing": 92.3, "mmlu_medical_genetics": 84.0, "mmlu_miscellaneous": 89.9, "mmlu_moral_disputes": 81.2, "mmlu_moral_scenarios": 68.2, "mmlu_nutrition": 82.7, "mmlu_philosophy": 77.5, "mmlu_prehistory": 85.2, "mmlu_pro": 52.8, "mmlu_professional_accounting": 63.1, "mmlu_professional_law": 57.0, "mmlu_professional_medicine": 81.2, "mmlu_professional_psychology": 78.3, "mmlu_public_relations": 72.7, "mmlu_security_studies": 82.0, "mmlu_sociology": 88.1, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 62.7, "mmlu_world_religions": 86.5, "musr": 18.2, "truthfulqa": 62.2, "winogrande": 81.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.6, "display_name": "Yi 1.5 34B Chat", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "01-ai/yi-1-5-34b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 65.8, "bbh": 48.2, "gpqa_diamond": 36.5, "gsm8k": 73.2, "hellaswag": 86.1, "ifeval": 77.3, "math_500": 58.5, "mmlu_abstract_algebra": 55.0, "mmlu_anatomy": 75.6, "mmlu_astronomy": 88.2, "mmlu_business_ethics": 80.0, "mmlu_clinical_knowledge": 83.0, "mmlu_college_biology": 87.5, "mmlu_college_chemistry": 57.0, "mmlu_college_computer_science": 70.0, "mmlu_college_mathematics": 59.0, "mmlu_college_medicine": 75.1, "mmlu_college_physics": 56.9, "mmlu_computer_security": 86.0, "mmlu_conceptual_physics": 83.8, "mmlu_econometrics": 64.9, "mmlu_electrical_engineering": 78.6, "mmlu_elementary_mathematics": 72.0, "mmlu_formal_logic": 67.5, "mmlu_global_facts": 55.0, "mmlu_high_school_biology": 87.7, "mmlu_high_school_chemistry": 69.5, "mmlu_high_school_computer_science": 83.0, "mmlu_high_school_european_history": 87.3, "mmlu_high_school_geography": 93.4, "mmlu_high_school_government_and_politics": 94.8, "mmlu_high_school_macroeconomics": 83.1, "mmlu_high_school_mathematics": 51.1, "mmlu_high_school_microeconomics": 88.7, "mmlu_high_school_physics": 55.6, "mmlu_high_school_psychology": 91.4, "mmlu_high_school_statistics": 72.7, "mmlu_high_school_us_history": 89.2, "mmlu_high_school_world_history": 90.7, "mmlu_human_aging": 81.2, "mmlu_human_sexuality": 87.8, "mmlu_international_law": 87.6, "mmlu_jurisprudence": 85.2, "mmlu_logical_fallacies": 87.1, "mmlu_machine_learning": 58.0, "mmlu_management": 91.3, "mmlu_marketing": 94.9, "mmlu_medical_genetics": 86.0, "mmlu_miscellaneous": 90.2, "mmlu_moral_disputes": 81.2, "mmlu_moral_scenarios": 64.7, "mmlu_nutrition": 84.3, "mmlu_philosophy": 81.4, "mmlu_prehistory": 84.6, "mmlu_pro": 52.8, "mmlu_professional_accounting": 62.1, "mmlu_professional_law": 58.0, "mmlu_professional_medicine": 84.9, "mmlu_professional_psychology": 80.9, "mmlu_public_relations": 76.4, "mmlu_security_studies": 82.4, "mmlu_sociology": 87.1, "mmlu_us_foreign_policy": 90.0, "mmlu_virology": 56.6, "mmlu_world_religions": 88.9, "musr": 18.2, "truthfulqa": 53.8, "winogrande": 84.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.6, "display_name": "Yi 1.5 34B", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "01-ai/yi-1-5-34b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 60.7, "bbh": 45.7, "gpqa_diamond": 30.2, "gsm8k": 67.1, "hellaswag": 78.9, "ifeval": 51.5, "math_500": 16.2, "mmlu_abstract_algebra": 45.0, "mmlu_anatomy": 51.9, "mmlu_astronomy": 67.1, "mmlu_business_ethics": 70.0, "mmlu_clinical_knowledge": 67.9, "mmlu_college_biology": 72.2, "mmlu_college_chemistry": 45.0, "mmlu_college_computer_science": 54.0, "mmlu_college_mathematics": 41.0, "mmlu_college_medicine": 60.7, "mmlu_college_physics": 45.1, "mmlu_computer_security": 71.0, "mmlu_conceptual_physics": 64.7, "mmlu_econometrics": 51.8, "mmlu_electrical_engineering": 63.4, "mmlu_elementary_mathematics": 53.4, "mmlu_formal_logic": 58.7, "mmlu_global_facts": 42.0, "mmlu_high_school_biology": 75.8, "mmlu_high_school_chemistry": 53.2, "mmlu_high_school_computer_science": 75.0, "mmlu_high_school_european_history": 78.8, "mmlu_high_school_geography": 78.3, "mmlu_high_school_government_and_politics": 84.5, "mmlu_high_school_macroeconomics": 66.9, "mmlu_high_school_mathematics": 42.6, "mmlu_high_school_microeconomics": 82.8, "mmlu_high_school_physics": 39.1, "mmlu_high_school_psychology": 82.4, "mmlu_high_school_statistics": 53.2, "mmlu_high_school_us_history": 77.0, "mmlu_high_school_world_history": 78.5, "mmlu_human_aging": 61.0, "mmlu_human_sexuality": 65.6, "mmlu_international_law": 76.0, "mmlu_jurisprudence": 77.8, "mmlu_logical_fallacies": 76.7, "mmlu_machine_learning": 56.2, "mmlu_management": 80.6, "mmlu_marketing": 88.9, "mmlu_medical_genetics": 68.0, "mmlu_miscellaneous": 79.7, "mmlu_moral_disputes": 69.7, "mmlu_moral_scenarios": 38.2, "mmlu_nutrition": 68.0, "mmlu_philosophy": 64.0, "mmlu_prehistory": 63.6, "mmlu_pro": 31.9, "mmlu_professional_accounting": 52.1, "mmlu_professional_law": 47.3, "mmlu_professional_medicine": 53.7, "mmlu_professional_psychology": 62.3, "mmlu_public_relations": 64.5, "mmlu_security_studies": 68.6, "mmlu_sociology": 81.1, "mmlu_us_foreign_policy": 81.0, "mmlu_virology": 50.0, "mmlu_world_religions": 74.3, "musr": 43.9, "truthfulqa": 52.6, "winogrande": 73.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 1.5 6B Chat", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "01-ai/yi-1-5-6b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 57.3, "bbh": 44.9, "gpqa_diamond": 31.4, "gsm8k": 49.8, "hellaswag": 78.0, "ifeval": 26.2, "math_500": 6.6, "mmlu_abstract_algebra": 37.0, "mmlu_anatomy": 55.6, "mmlu_astronomy": 66.4, "mmlu_business_ethics": 70.0, "mmlu_clinical_knowledge": 65.7, "mmlu_college_biology": 70.8, "mmlu_college_chemistry": 44.0, "mmlu_college_computer_science": 57.0, "mmlu_college_mathematics": 39.0, "mmlu_college_medicine": 60.7, "mmlu_college_physics": 44.1, "mmlu_computer_security": 79.0, "mmlu_conceptual_physics": 66.8, "mmlu_econometrics": 46.5, "mmlu_electrical_engineering": 65.5, "mmlu_elementary_mathematics": 51.9, "mmlu_formal_logic": 49.2, "mmlu_global_facts": 34.0, "mmlu_high_school_biology": 77.7, "mmlu_high_school_chemistry": 62.1, "mmlu_high_school_computer_science": 69.0, "mmlu_high_school_european_history": 79.4, "mmlu_high_school_geography": 80.8, "mmlu_high_school_government_and_politics": 87.6, "mmlu_high_school_macroeconomics": 71.0, "mmlu_high_school_mathematics": 40.7, "mmlu_high_school_microeconomics": 81.9, "mmlu_high_school_physics": 44.4, "mmlu_high_school_psychology": 83.1, "mmlu_high_school_statistics": 57.4, "mmlu_high_school_us_history": 82.4, "mmlu_high_school_world_history": 77.2, "mmlu_human_aging": 62.8, "mmlu_human_sexuality": 70.2, "mmlu_international_law": 75.2, "mmlu_jurisprudence": 77.8, "mmlu_logical_fallacies": 73.6, "mmlu_machine_learning": 46.4, "mmlu_management": 79.6, "mmlu_marketing": 90.2, "mmlu_medical_genetics": 75.0, "mmlu_miscellaneous": 80.6, "mmlu_moral_disputes": 69.4, "mmlu_moral_scenarios": 33.2, "mmlu_nutrition": 69.3, "mmlu_philosophy": 68.5, "mmlu_prehistory": 71.3, "mmlu_pro": 31.4, "mmlu_professional_accounting": 54.3, "mmlu_professional_law": 47.1, "mmlu_professional_medicine": 65.8, "mmlu_professional_psychology": 65.5, "mmlu_public_relations": 69.1, "mmlu_security_studies": 72.7, "mmlu_sociology": 77.1, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 50.6, "mmlu_world_religions": 76.0, "musr": 43.7, "truthfulqa": 44.0, "winogrande": 75.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 1.5 6B", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "01-ai/yi-1-5-6b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 61.1, "bbh": 49.6, "gpqa_diamond": 35.9, "gsm8k": 0.0, "hellaswag": 79.8, "ifeval": 23.0, "math_500": 10.8, "mmlu_abstract_algebra": 48.0, "mmlu_anatomy": 65.9, "mmlu_astronomy": 77.0, "mmlu_business_ethics": 75.0, "mmlu_clinical_knowledge": 71.7, "mmlu_college_biology": 81.9, "mmlu_college_chemistry": 49.0, "mmlu_college_computer_science": 60.0, "mmlu_college_mathematics": 53.0, "mmlu_college_medicine": 70.5, "mmlu_college_physics": 47.1, "mmlu_computer_security": 84.0, "mmlu_conceptual_physics": 71.9, "mmlu_econometrics": 63.2, "mmlu_electrical_engineering": 71.0, "mmlu_elementary_mathematics": 60.3, "mmlu_formal_logic": 57.1, "mmlu_global_facts": 43.0, "mmlu_high_school_biology": 84.2, "mmlu_high_school_chemistry": 63.5, "mmlu_high_school_computer_science": 82.0, "mmlu_high_school_european_history": 82.4, "mmlu_high_school_geography": 84.8, "mmlu_high_school_government_and_politics": 91.2, "mmlu_high_school_macroeconomics": 73.6, "mmlu_high_school_mathematics": 43.0, "mmlu_high_school_microeconomics": 82.8, "mmlu_high_school_physics": 46.4, "mmlu_high_school_psychology": 86.4, "mmlu_high_school_statistics": 66.2, "mmlu_high_school_us_history": 84.3, "mmlu_high_school_world_history": 84.4, "mmlu_human_aging": 74.4, "mmlu_human_sexuality": 80.2, "mmlu_international_law": 81.0, "mmlu_jurisprudence": 80.6, "mmlu_logical_fallacies": 74.2, "mmlu_machine_learning": 50.9, "mmlu_management": 81.6, "mmlu_marketing": 89.7, "mmlu_medical_genetics": 80.0, "mmlu_miscellaneous": 84.4, "mmlu_moral_disputes": 75.7, "mmlu_moral_scenarios": 45.8, "mmlu_nutrition": 78.1, "mmlu_philosophy": 76.8, "mmlu_prehistory": 74.4, "mmlu_pro": 37.6, "mmlu_professional_accounting": 56.7, "mmlu_professional_law": 50.7, "mmlu_professional_medicine": 69.5, "mmlu_professional_psychology": 71.7, "mmlu_public_relations": 73.6, "mmlu_security_studies": 75.9, "mmlu_sociology": 83.6, "mmlu_us_foreign_policy": 93.0, "mmlu_virology": 54.2, "mmlu_world_religions": 83.6, "musr": 41.9, "truthfulqa": 42.5, "winogrande": 76.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.15, "display_name": "Yi 1.5 9B 32K", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-1-5-9b-32k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 64.3, "bbh": 51.5, "gpqa_diamond": 30.9, "gsm8k": 59.7, "hellaswag": 80.9, "ifeval": 42.1, "math_500": 17.8, "mmlu_abstract_algebra": 46.0, "mmlu_anatomy": 68.1, "mmlu_astronomy": 74.3, "mmlu_business_ethics": 76.0, "mmlu_clinical_knowledge": 72.5, "mmlu_college_biology": 78.5, "mmlu_college_chemistry": 54.0, "mmlu_college_computer_science": 62.0, "mmlu_college_mathematics": 47.0, "mmlu_college_medicine": 69.4, "mmlu_college_physics": 47.1, "mmlu_computer_security": 77.0, "mmlu_conceptual_physics": 71.1, "mmlu_econometrics": 59.6, "mmlu_electrical_engineering": 66.2, "mmlu_elementary_mathematics": 64.3, "mmlu_formal_logic": 61.1, "mmlu_global_facts": 45.0, "mmlu_high_school_biology": 84.5, "mmlu_high_school_chemistry": 61.1, "mmlu_high_school_computer_science": 87.0, "mmlu_high_school_european_history": 83.6, "mmlu_high_school_geography": 83.3, "mmlu_high_school_government_and_politics": 89.1, "mmlu_high_school_macroeconomics": 76.2, "mmlu_high_school_mathematics": 45.6, "mmlu_high_school_microeconomics": 84.5, "mmlu_high_school_physics": 51.0, "mmlu_high_school_psychology": 85.7, "mmlu_high_school_statistics": 65.3, "mmlu_high_school_us_history": 86.8, "mmlu_high_school_world_history": 84.8, "mmlu_human_aging": 71.7, "mmlu_human_sexuality": 74.8, "mmlu_international_law": 80.2, "mmlu_jurisprudence": 77.8, "mmlu_logical_fallacies": 82.2, "mmlu_machine_learning": 51.8, "mmlu_management": 76.7, "mmlu_marketing": 91.5, "mmlu_medical_genetics": 80.0, "mmlu_miscellaneous": 84.2, "mmlu_moral_disputes": 72.3, "mmlu_moral_scenarios": 54.6, "mmlu_nutrition": 77.5, "mmlu_philosophy": 75.9, "mmlu_prehistory": 74.1, "mmlu_pro": 39.9, "mmlu_professional_accounting": 57.8, "mmlu_professional_law": 50.1, "mmlu_professional_medicine": 68.0, "mmlu_professional_psychology": 70.8, "mmlu_public_relations": 72.7, "mmlu_security_studies": 77.1, "mmlu_sociology": 81.1, "mmlu_us_foreign_policy": 93.0, "mmlu_virology": 57.2, "mmlu_world_religions": 84.2, "musr": 41.0, "truthfulqa": 51.0, "winogrande": 75.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 16384, "cost_input": 0.15, "display_name": "Yi 1.5 9B Chat 16K", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-1-5-9b-chat-16k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 63.7, "bbh": 55.6, "gpqa_diamond": 33.5, "gsm8k": 71.9, "hellaswag": 80.9, "ifeval": 60.5, "math_500": 22.6, "mmlu_abstract_algebra": 46.0, "mmlu_anatomy": 63.7, "mmlu_astronomy": 72.4, "mmlu_business_ethics": 74.0, "mmlu_clinical_knowledge": 72.8, "mmlu_college_biology": 79.2, "mmlu_college_chemistry": 59.0, "mmlu_college_computer_science": 66.0, "mmlu_college_mathematics": 48.0, "mmlu_college_medicine": 71.1, "mmlu_college_physics": 46.1, "mmlu_computer_security": 77.0, "mmlu_conceptual_physics": 72.3, "mmlu_econometrics": 58.8, "mmlu_electrical_engineering": 69.7, "mmlu_elementary_mathematics": 64.0, "mmlu_formal_logic": 57.9, "mmlu_global_facts": 45.0, "mmlu_high_school_biology": 83.5, "mmlu_high_school_chemistry": 66.5, "mmlu_high_school_computer_science": 87.0, "mmlu_high_school_european_history": 84.2, "mmlu_high_school_geography": 85.9, "mmlu_high_school_government_and_politics": 90.2, "mmlu_high_school_macroeconomics": 77.9, "mmlu_high_school_mathematics": 47.0, "mmlu_high_school_microeconomics": 87.0, "mmlu_high_school_physics": 45.7, "mmlu_high_school_psychology": 86.4, "mmlu_high_school_statistics": 68.5, "mmlu_high_school_us_history": 84.8, "mmlu_high_school_world_history": 82.7, "mmlu_human_aging": 72.2, "mmlu_human_sexuality": 74.8, "mmlu_international_law": 83.5, "mmlu_jurisprudence": 75.0, "mmlu_logical_fallacies": 80.4, "mmlu_machine_learning": 53.6, "mmlu_management": 86.4, "mmlu_marketing": 91.9, "mmlu_medical_genetics": 80.0, "mmlu_miscellaneous": 83.4, "mmlu_moral_disputes": 73.7, "mmlu_moral_scenarios": 55.0, "mmlu_nutrition": 74.5, "mmlu_philosophy": 73.6, "mmlu_prehistory": 73.1, "mmlu_pro": 39.8, "mmlu_professional_accounting": 57.8, "mmlu_professional_law": 48.0, "mmlu_professional_medicine": 67.6, "mmlu_professional_psychology": 69.6, "mmlu_public_relations": 69.1, "mmlu_security_studies": 73.1, "mmlu_sociology": 81.6, "mmlu_us_foreign_policy": 91.0, "mmlu_virology": 56.0, "mmlu_world_religions": 82.5, "musr": 42.6, "truthfulqa": 52.7, "winogrande": 77.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.15, "display_name": "Yi 1.5 9B Chat", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-1-5-9b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 61.9, "bbh": 51.4, "gpqa_diamond": 37.9, "gsm8k": 62.7, "hellaswag": 80.4, "ifeval": 29.4, "math_500": 11.4, "mmlu_abstract_algebra": 44.0, "mmlu_anatomy": 66.7, "mmlu_astronomy": 75.7, "mmlu_business_ethics": 78.0, "mmlu_clinical_knowledge": 68.3, "mmlu_college_biology": 79.2, "mmlu_college_chemistry": 53.0, "mmlu_college_computer_science": 58.0, "mmlu_college_mathematics": 53.0, "mmlu_college_medicine": 69.9, "mmlu_college_physics": 47.1, "mmlu_computer_security": 85.0, "mmlu_conceptual_physics": 73.6, "mmlu_econometrics": 63.2, "mmlu_electrical_engineering": 69.0, "mmlu_elementary_mathematics": 63.2, "mmlu_formal_logic": 53.2, "mmlu_global_facts": 40.0, "mmlu_high_school_biology": 84.5, "mmlu_high_school_chemistry": 63.5, "mmlu_high_school_computer_science": 83.0, "mmlu_high_school_european_history": 83.0, "mmlu_high_school_geography": 84.8, "mmlu_high_school_government_and_politics": 90.2, "mmlu_high_school_macroeconomics": 75.4, "mmlu_high_school_mathematics": 43.7, "mmlu_high_school_microeconomics": 86.1, "mmlu_high_school_physics": 47.7, "mmlu_high_school_psychology": 85.9, "mmlu_high_school_statistics": 65.3, "mmlu_high_school_us_history": 84.3, "mmlu_high_school_world_history": 84.8, "mmlu_human_aging": 74.4, "mmlu_human_sexuality": 77.9, "mmlu_international_law": 82.6, "mmlu_jurisprudence": 74.1, "mmlu_logical_fallacies": 78.5, "mmlu_machine_learning": 50.0, "mmlu_management": 80.6, "mmlu_marketing": 89.7, "mmlu_medical_genetics": 77.0, "mmlu_miscellaneous": 83.0, "mmlu_moral_disputes": 74.6, "mmlu_moral_scenarios": 48.6, "mmlu_nutrition": 78.4, "mmlu_philosophy": 75.6, "mmlu_prehistory": 75.3, "mmlu_pro": 39.2, "mmlu_professional_accounting": 56.7, "mmlu_professional_law": 50.3, "mmlu_professional_medicine": 70.2, "mmlu_professional_psychology": 73.9, "mmlu_public_relations": 70.9, "mmlu_security_studies": 75.5, "mmlu_sociology": 84.1, "mmlu_us_foreign_policy": 91.0, "mmlu_virology": 53.6, "mmlu_world_religions": 83.0, "musr": 43.3, "truthfulqa": 46.7, "winogrande": 78.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.15, "display_name": "Yi 1.5 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-1-5-9b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 65.8, "bbh": 54.4, "gpqa_diamond": 35.7, "gsm8k": 34.9, "hellaswag": 82.1, "ifeval": 15.4, "math_500": 5.7, "mmlu_abstract_algebra": 42.0, "mmlu_anatomy": 73.3, "mmlu_astronomy": 89.5, "mmlu_business_ethics": 78.0, "mmlu_clinical_knowledge": 81.1, "mmlu_college_biology": 88.2, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 63.0, "mmlu_college_mathematics": 40.0, "mmlu_college_medicine": 71.1, "mmlu_college_physics": 54.9, "mmlu_computer_security": 80.0, "mmlu_conceptual_physics": 77.4, "mmlu_econometrics": 56.1, "mmlu_electrical_engineering": 77.2, "mmlu_elementary_mathematics": 64.6, "mmlu_formal_logic": 51.6, "mmlu_global_facts": 57.0, "mmlu_high_school_biology": 89.7, "mmlu_high_school_chemistry": 65.0, "mmlu_high_school_computer_science": 82.0, "mmlu_high_school_european_history": 84.8, "mmlu_high_school_geography": 92.9, "mmlu_high_school_government_and_politics": 97.4, "mmlu_high_school_macroeconomics": 78.2, "mmlu_high_school_mathematics": 38.1, "mmlu_high_school_microeconomics": 84.0, "mmlu_high_school_physics": 46.4, "mmlu_high_school_psychology": 92.1, "mmlu_high_school_statistics": 67.1, "mmlu_high_school_us_history": 90.7, "mmlu_high_school_world_history": 90.7, "mmlu_human_aging": 79.4, "mmlu_human_sexuality": 86.3, "mmlu_international_law": 91.7, "mmlu_jurisprudence": 88.9, "mmlu_logical_fallacies": 88.3, "mmlu_machine_learning": 58.0, "mmlu_management": 89.3, "mmlu_marketing": 94.0, "mmlu_medical_genetics": 88.0, "mmlu_miscellaneous": 90.3, "mmlu_moral_disputes": 80.3, "mmlu_moral_scenarios": 67.6, "mmlu_nutrition": 84.6, "mmlu_philosophy": 82.0, "mmlu_prehistory": 83.6, "mmlu_pro": 45.3, "mmlu_professional_accounting": 61.3, "mmlu_professional_law": 59.1, "mmlu_professional_medicine": 80.9, "mmlu_professional_psychology": 82.0, "mmlu_public_relations": 72.7, "mmlu_security_studies": 82.0, "mmlu_sociology": 88.1, "mmlu_us_foreign_policy": 92.0, "mmlu_virology": 57.2, "mmlu_world_religions": 87.7, "musr": 38.2, "truthfulqa": 42.6, "winogrande": 82.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": null, "display_name": "Yi 34B 200K", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "01-ai/yi-34b-200k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 65.4, "bbh": 55.6, "gpqa_diamond": 33.8, "gsm8k": 31.9, "hellaswag": 84.2, "ifeval": 47.0, "math_500": 6.3, "mmlu_abstract_algebra": 45.0, "mmlu_anatomy": 71.1, "mmlu_astronomy": 85.5, "mmlu_business_ethics": 79.0, "mmlu_clinical_knowledge": 79.2, "mmlu_college_biology": 84.7, "mmlu_college_chemistry": 53.0, "mmlu_college_computer_science": 59.0, "mmlu_college_mathematics": 34.0, "mmlu_college_medicine": 69.9, "mmlu_college_physics": 46.1, "mmlu_computer_security": 82.0, "mmlu_conceptual_physics": 76.6, "mmlu_econometrics": 55.3, "mmlu_electrical_engineering": 80.0, "mmlu_elementary_mathematics": 63.5, "mmlu_formal_logic": 54.0, "mmlu_global_facts": 56.0, "mmlu_high_school_biology": 87.1, "mmlu_high_school_chemistry": 62.1, "mmlu_high_school_computer_science": 81.0, "mmlu_high_school_european_history": 85.5, "mmlu_high_school_geography": 89.9, "mmlu_high_school_government_and_politics": 95.3, "mmlu_high_school_macroeconomics": 78.5, "mmlu_high_school_mathematics": 35.6, "mmlu_high_school_microeconomics": 83.6, "mmlu_high_school_physics": 50.3, "mmlu_high_school_psychology": 90.8, "mmlu_high_school_statistics": 63.4, "mmlu_high_school_us_history": 90.7, "mmlu_high_school_world_history": 90.3, "mmlu_human_aging": 81.2, "mmlu_human_sexuality": 89.3, "mmlu_international_law": 89.3, "mmlu_jurisprudence": 88.9, "mmlu_logical_fallacies": 85.3, "mmlu_machine_learning": 61.6, "mmlu_management": 87.4, "mmlu_marketing": 91.5, "mmlu_medical_genetics": 87.0, "mmlu_miscellaneous": 89.9, "mmlu_moral_disputes": 80.9, "mmlu_moral_scenarios": 70.2, "mmlu_nutrition": 83.0, "mmlu_philosophy": 81.0, "mmlu_prehistory": 87.0, "mmlu_pro": 40.9, "mmlu_professional_accounting": 61.7, "mmlu_professional_law": 55.0, "mmlu_professional_medicine": 78.3, "mmlu_professional_psychology": 82.7, "mmlu_public_relations": 72.7, "mmlu_security_studies": 83.7, "mmlu_sociology": 88.6, "mmlu_us_foreign_policy": 89.0, "mmlu_virology": 58.4, "mmlu_world_religions": 87.7, "musr": 39.8, "truthfulqa": 55.4, "winogrande": 80.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 34B Chat", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "01-ai/yi-34b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 54.6, "gpqa_diamond": 36.7, "gsm8k": 50.6, "ifeval": 30.5, "math_500": 5.1, "mmlu_pro": 44.1, "musr": 41.2, "winogrande": 83.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 34B", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "01-ai/yi-34b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 42.9, "gpqa_diamond": 28.2, "gsm8k": 30.3, "ifeval": 8.4, "math_500": 1.8, "mmlu_pro": 28.4, "musr": 45.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": null, "display_name": "Yi 6B 200K", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "01-ai/yi-6b-200k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 41.3, "gpqa_diamond": 29.4, "ifeval": 34.0, "math_500": 1.4, "mmlu_pro": 30.6, "musr": 36.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 6B Chat 4bits", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "01-ai/yi-6b-chat-4bits", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"arc_challenge": 55.5, "bbh": 41.3, "gpqa_diamond": 29.4, "gsm8k": 12.1, "hellaswag": 76.6, "ifeval": 34.0, "math_500": 1.4, "mmlu_abstract_algebra": 30.0, "mmlu_anatomy": 61.5, "mmlu_astronomy": 65.8, "mmlu_business_ethics": 67.0, "mmlu_clinical_knowledge": 66.8, "mmlu_college_biology": 66.0, "mmlu_college_chemistry": 43.0, "mmlu_college_computer_science": 51.0, "mmlu_college_mathematics": 42.0, "mmlu_college_medicine": 60.7, "mmlu_college_physics": 34.3, "mmlu_computer_security": 75.0, "mmlu_conceptual_physics": 60.4, "mmlu_econometrics": 35.1, "mmlu_electrical_engineering": 64.1, "mmlu_elementary_mathematics": 45.0, "mmlu_formal_logic": 43.7, "mmlu_global_facts": 45.0, "mmlu_high_school_biology": 77.4, "mmlu_high_school_chemistry": 53.2, "mmlu_high_school_computer_science": 65.0, "mmlu_high_school_european_history": 75.2, "mmlu_high_school_geography": 82.3, "mmlu_high_school_government_and_politics": 89.6, "mmlu_high_school_macroeconomics": 64.9, "mmlu_high_school_mathematics": 32.6, "mmlu_high_school_microeconomics": 73.9, "mmlu_high_school_physics": 39.1, "mmlu_high_school_psychology": 82.8, "mmlu_high_school_statistics": 51.9, "mmlu_high_school_us_history": 80.9, "mmlu_high_school_world_history": 78.9, "mmlu_human_aging": 67.3, "mmlu_human_sexuality": 74.0, "mmlu_international_law": 76.9, "mmlu_jurisprudence": 79.6, "mmlu_logical_fallacies": 79.1, "mmlu_machine_learning": 42.0, "mmlu_management": 80.6, "mmlu_marketing": 90.2, "mmlu_medical_genetics": 79.0, "mmlu_miscellaneous": 80.7, "mmlu_moral_disputes": 69.9, "mmlu_moral_scenarios": 42.5, "mmlu_nutrition": 71.6, "mmlu_philosophy": 69.1, "mmlu_prehistory": 71.0, "mmlu_pro": 30.6, "mmlu_professional_accounting": 47.9, "mmlu_professional_law": 49.5, "mmlu_professional_medicine": 67.3, "mmlu_professional_psychology": 66.0, "mmlu_public_relations": 73.6, "mmlu_security_studies": 75.9, "mmlu_sociology": 82.1, "mmlu_us_foreign_policy": 85.0, "mmlu_virology": 46.4, "mmlu_world_religions": 83.0, "musr": 36.9, "truthfulqa": 42.0, "winogrande": 74.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 6B Chat", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "01-ai/yi-6b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 55.5, "bbh": 43.1, "gpqa_diamond": 26.9, "gsm8k": 12.1, "hellaswag": 76.6, "ifeval": 28.9, "math_500": 1.6, "mmlu_abstract_algebra": 30.0, "mmlu_anatomy": 61.5, "mmlu_astronomy": 65.8, "mmlu_business_ethics": 67.0, "mmlu_clinical_knowledge": 66.8, "mmlu_college_biology": 66.0, "mmlu_college_chemistry": 43.0, "mmlu_college_computer_science": 51.0, "mmlu_college_mathematics": 42.0, "mmlu_college_medicine": 60.7, "mmlu_college_physics": 34.3, "mmlu_computer_security": 75.0, "mmlu_conceptual_physics": 60.4, "mmlu_econometrics": 35.1, "mmlu_electrical_engineering": 64.1, "mmlu_elementary_mathematics": 45.0, "mmlu_formal_logic": 43.7, "mmlu_global_facts": 45.0, "mmlu_high_school_biology": 77.4, "mmlu_high_school_chemistry": 53.2, "mmlu_high_school_computer_science": 65.0, "mmlu_high_school_european_history": 75.2, "mmlu_high_school_geography": 82.3, "mmlu_high_school_government_and_politics": 89.6, "mmlu_high_school_macroeconomics": 64.9, "mmlu_high_school_mathematics": 32.6, "mmlu_high_school_microeconomics": 73.9, "mmlu_high_school_physics": 39.1, "mmlu_high_school_psychology": 82.8, "mmlu_high_school_statistics": 51.9, "mmlu_high_school_us_history": 80.9, "mmlu_high_school_world_history": 78.9, "mmlu_human_aging": 67.3, "mmlu_human_sexuality": 74.0, "mmlu_international_law": 76.9, "mmlu_jurisprudence": 79.6, "mmlu_logical_fallacies": 79.1, "mmlu_machine_learning": 42.0, "mmlu_management": 80.6, "mmlu_marketing": 90.2, "mmlu_medical_genetics": 79.0, "mmlu_miscellaneous": 80.7, "mmlu_moral_disputes": 69.9, "mmlu_moral_scenarios": 42.5, "mmlu_nutrition": 71.6, "mmlu_philosophy": 69.1, "mmlu_prehistory": 71.0, "mmlu_pro": 29.9, "mmlu_professional_accounting": 47.9, "mmlu_professional_law": 49.5, "mmlu_professional_medicine": 67.3, "mmlu_professional_psychology": 66.0, "mmlu_public_relations": 73.6, "mmlu_security_studies": 75.9, "mmlu_sociology": 82.1, "mmlu_us_foreign_policy": 85.0, "mmlu_virology": 46.4, "mmlu_world_religions": 83.0, "musr": 39.4, "truthfulqa": 42.0, "winogrande": 74.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 6B", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "01-ai/yi-6b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 47.9, "gpqa_diamond": 31.5, "ifeval": 23.3, "math_500": 6.6, "mmlu_pro": 36.2, "musr": 42.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": null, "display_name": "Yi 9B 200K", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-9b-200k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"arc_challenge": 61.2, "bbh": 49.4, "gpqa_diamond": 31.8, "gsm8k": 49.0, "hellaswag": 78.8, "ifeval": 27.1, "math_500": 5.6, "mmlu_abstract_algebra": 34.0, "mmlu_anatomy": 57.8, "mmlu_astronomy": 75.7, "mmlu_business_ethics": 80.0, "mmlu_clinical_knowledge": 73.2, "mmlu_college_biology": 82.6, "mmlu_college_chemistry": 54.0, "mmlu_college_computer_science": 62.0, "mmlu_college_mathematics": 43.0, "mmlu_college_medicine": 74.6, "mmlu_college_physics": 42.2, "mmlu_computer_security": 82.0, "mmlu_conceptual_physics": 71.9, "mmlu_econometrics": 52.6, "mmlu_electrical_engineering": 72.4, "mmlu_elementary_mathematics": 60.1, "mmlu_formal_logic": 58.7, "mmlu_global_facts": 33.0, "mmlu_high_school_biology": 84.5, "mmlu_high_school_chemistry": 58.1, "mmlu_high_school_computer_science": 83.0, "mmlu_high_school_european_history": 78.2, "mmlu_high_school_geography": 85.4, "mmlu_high_school_government_and_politics": 93.8, "mmlu_high_school_macroeconomics": 75.9, "mmlu_high_school_mathematics": 40.0, "mmlu_high_school_microeconomics": 82.4, "mmlu_high_school_physics": 41.1, "mmlu_high_school_psychology": 86.8, "mmlu_high_school_statistics": 68.5, "mmlu_high_school_us_history": 86.3, "mmlu_high_school_world_history": 81.9, "mmlu_human_aging": 74.4, "mmlu_human_sexuality": 78.6, "mmlu_international_law": 83.5, "mmlu_jurisprudence": 79.6, "mmlu_logical_fallacies": 76.1, "mmlu_machine_learning": 58.0, "mmlu_management": 83.5, "mmlu_marketing": 91.0, "mmlu_medical_genetics": 75.0, "mmlu_miscellaneous": 83.9, "mmlu_moral_disputes": 76.6, "mmlu_moral_scenarios": 40.0, "mmlu_nutrition": 75.5, "mmlu_philosophy": 77.5, "mmlu_prehistory": 76.9, "mmlu_pro": 35.7, "mmlu_professional_accounting": 56.4, "mmlu_professional_law": 49.5, "mmlu_professional_medicine": 71.0, "mmlu_professional_psychology": 70.4, "mmlu_public_relations": 74.5, "mmlu_security_studies": 75.9, "mmlu_sociology": 86.1, "mmlu_us_foreign_policy": 90.0, "mmlu_virology": 51.2, "mmlu_world_religions": 83.0, "musr": 40.5, "truthfulqa": 42.4, "winogrande": 77.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Yi 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-9b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "01.AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 48.1, "gpqa_diamond": 24.7, "ifeval": 48.2, "math_500": 4.0, "mmlu_pro": 24.3, "musr": 39.9}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Yi Coder 9B Chat", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-coder-9b-chat", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "01.AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 48.1, "gpqa_diamond": 24.7, "ifeval": 48.2, "math_500": 4.0, "mmlu_pro": 24.3, "musr": 39.9}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Yi Coder 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "01-ai/yi-coder-9b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "01.AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba Large 1.5", "fits": {"apple_m3_ultra": 2.9}, "model_id": "ai21/ai21-jamba-large-1-5", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba Large 1.6", "fits": {"apple_m3_ultra": 2.9}, "model_id": "ai21/ai21-jamba-large-1-6", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba Mini 1.5", "fits": {"apple_m1_max": 10.8, "apple_m2_max": 10.8, "apple_m2_ultra": 21.5, "apple_m3_max": 10.8, "apple_m3_ultra": 22.0, "apple_m4_max": 14.7, "apple_m4_pro": 7.3}, "model_id": "ai21/ai21-jamba-mini-1-5", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba Mini 1.6", "fits": {"apple_m1_max": 10.8, "apple_m2_max": 10.8, "apple_m2_ultra": 21.5, "apple_m3_max": 10.8, "apple_m3_ultra": 22.0, "apple_m4_max": 14.7, "apple_m4_pro": 7.3}, "model_id": "ai21/ai21-jamba-mini-1-6", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba Mini 1.7", "fits": {"apple_m1_max": 10.8, "apple_m2_max": 10.8, "apple_m2_ultra": 21.5, "apple_m3_max": 10.8, "apple_m3_ultra": 22.0, "apple_m4_max": 14.7, "apple_m4_pro": 7.3}, "model_id": "ai21/ai21-jamba-mini-1-7", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba Reasoning 3B GGUF", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "ai21/ai21-jamba-reasoning-3b-gguf", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba Reasoning 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "ai21/ai21-jamba-reasoning-3b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba2 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "ai21/ai21-jamba2-3b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba2 Mini FP8", "fits": {"apple_m1_max": 10.8, "apple_m2_max": 10.8, "apple_m2_ultra": 21.5, "apple_m3_max": 10.8, "apple_m3_ultra": 22.0, "apple_m4_max": 14.7, "apple_m4_pro": 7.3}, "model_id": "ai21/ai21-jamba2-mini-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "AI21 Jamba2 Mini", "fits": {"apple_m1_max": 10.8, "apple_m2_max": 10.8, "apple_m2_ultra": 21.5, "apple_m3_max": 10.8, "apple_m3_ultra": 22.0, "apple_m4_max": 14.7, "apple_m4_pro": 7.3}, "model_id": "ai21/ai21-jamba2-mini", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "Jamba tiny dev", "fits": {"amd_rx_7900_xt": 5600.0, "amd_rx_7900_xtx": 6720.0, "amd_rx_9070_xt": 4515.0, "apple_m1_max": 2800.0, "apple_m2_max": 2800.0, "apple_m2_ultra": 5600.0, "apple_m3_max": 2800.0, "apple_m3_ultra": 5733.0, "apple_m4": 840.0, "apple_m4_max": 3822.0, "apple_m4_pro": 1911.0, "nvidia_rtx_3060_12gb": 2520.0, "nvidia_rtx_3090": 6552.0, "nvidia_rtx_4060_ti_16gb": 2016.0, "nvidia_rtx_4070_ti_super": 4704.0, "nvidia_rtx_4080_super": 5152.0, "nvidia_rtx_4090": 7056.0, "nvidia_rtx_5080": 6720.0, "nvidia_rtx_5090": 12544.0}, "model_id": "ai21/jamba-tiny-dev", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "Jamba tiny random", "fits": {"amd_rx_7900_xt": 5600.0, "amd_rx_7900_xtx": 6720.0, "amd_rx_9070_xt": 4515.0, "apple_m1_max": 2800.0, "apple_m2_max": 2800.0, "apple_m2_ultra": 5600.0, "apple_m3_max": 2800.0, "apple_m3_ultra": 5733.0, "apple_m4": 840.0, "apple_m4_max": 3822.0, "apple_m4_pro": 1911.0, "nvidia_rtx_3060_12gb": 2520.0, "nvidia_rtx_3090": 6552.0, "nvidia_rtx_4060_ti_16gb": 2016.0, "nvidia_rtx_4070_ti_super": 4704.0, "nvidia_rtx_4080_super": 5152.0, "nvidia_rtx_4090": 7056.0, "nvidia_rtx_5080": 6720.0, "nvidia_rtx_5090": 12544.0}, "model_id": "ai21/jamba-tiny-random", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Jamba tiny reward dev", "fits": {"amd_rx_7900_xt": 5600.0, "amd_rx_7900_xtx": 6720.0, "amd_rx_9070_xt": 4515.0, "apple_m1_max": 2800.0, "apple_m2_max": 2800.0, "apple_m2_ultra": 5600.0, "apple_m3_max": 2800.0, "apple_m3_ultra": 5733.0, "apple_m4": 840.0, "apple_m4_max": 3822.0, "apple_m4_pro": 1911.0, "nvidia_rtx_3060_12gb": 2520.0, "nvidia_rtx_3090": 6552.0, "nvidia_rtx_4060_ti_16gb": 2016.0, "nvidia_rtx_4070_ti_super": 4704.0, "nvidia_rtx_4080_super": 5152.0, "nvidia_rtx_4090": 7056.0, "nvidia_rtx_5080": 6720.0, "nvidia_rtx_5090": 12544.0}, "model_id": "ai21/jamba-tiny-reward-dev", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": null}, {"benchmark_scores": {"bbh": 36.0, "gpqa_diamond": 26.8, "ifeval": 20.3, "math_500": 1.6, "mmlu_pro": 24.9, "musr": 35.9}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 256000, "cost_input": null, "display_name": "Jamba v0.1", "fits": {"apple_m1_max": 10.8, "apple_m2_max": 10.8, "apple_m2_ultra": 21.5, "apple_m3_max": 10.8, "apple_m3_ultra": 22.0, "apple_m4_max": 14.7, "apple_m4_pro": 7.3}, "model_id": "ai21/jamba-v0-1", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "AI21 Labs", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "biomed roberta base", "fits": {"amd_rx_7900_xt": 8960.0, "amd_rx_7900_xtx": 10752.0, "amd_rx_9070_xt": 7224.0, "apple_m1_max": 4480.0, "apple_m2_max": 4480.0, "apple_m2_ultra": 8960.0, "apple_m3_max": 4480.0, "apple_m3_ultra": 9172.8, "apple_m4": 1344.0, "apple_m4_max": 6115.2, "apple_m4_pro": 3057.6, "nvidia_rtx_3060_12gb": 4032.0, "nvidia_rtx_3090": 10483.2, "nvidia_rtx_4060_ti_16gb": 3225.6, "nvidia_rtx_4070_ti_super": 7526.4, "nvidia_rtx_4080_super": 8243.2, "nvidia_rtx_4090": 11289.6, "nvidia_rtx_5080": 10752.0, "nvidia_rtx_5090": 20070.4}, "model_id": "allen-ai/biomed-roberta-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 16384, "cost_input": null, "display_name": "led base 16384", "fits": {"amd_rx_7900_xt": 6913.6, "amd_rx_7900_xtx": 8296.3, "amd_rx_9070_xt": 5574.1, "apple_m1_max": 3456.8, "apple_m2_max": 3456.8, "apple_m2_ultra": 6913.6, "apple_m3_max": 3456.8, "apple_m3_ultra": 7077.8, "apple_m4": 1037.0, "apple_m4_max": 4718.5, "apple_m4_pro": 2359.3, "nvidia_rtx_3060_12gb": 3111.1, "nvidia_rtx_3090": 8088.9, "nvidia_rtx_4060_ti_16gb": 2488.9, "nvidia_rtx_4070_ti_super": 5807.4, "nvidia_rtx_4080_super": 6360.5, "nvidia_rtx_4090": 8711.1, "nvidia_rtx_5080": 8296.3, "nvidia_rtx_5090": 15486.4}, "model_id": "allen-ai/led-base-16384", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "longformer base 4096", "fits": {"amd_rx_7900_xt": 7516.8, "amd_rx_7900_xtx": 9020.1, "amd_rx_9070_xt": 6060.4, "apple_m1_max": 3758.4, "apple_m2_max": 3758.4, "apple_m2_ultra": 7516.8, "apple_m3_max": 3758.4, "apple_m3_ultra": 7695.3, "apple_m4": 1127.5, "apple_m4_max": 5130.2, "apple_m4_pro": 2565.1, "nvidia_rtx_3060_12gb": 3382.6, "nvidia_rtx_3090": 8794.6, "nvidia_rtx_4060_ti_16gb": 2706.0, "nvidia_rtx_4070_ti_super": 6314.1, "nvidia_rtx_4080_super": 6915.4, "nvidia_rtx_4090": 9471.1, "nvidia_rtx_5080": 9020.1, "nvidia_rtx_5090": 16837.6}, "model_id": "allen-ai/longformer-base-4096", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "Molmo 7B D 0924", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/molmo-7b-d-0924", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "Molmo2 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "allen-ai/molmo2-4b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "Molmo2 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "allen-ai/molmo2-8b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "Molmo2 O 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/molmo2-o-7b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "Molmo2 VideoPoint 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "allen-ai/molmo2-videopoint-4b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "MolmoPoint Vid 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "allen-ai/molmopoint-vid-4b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 34.6, "bbh": 30.5, "gpqa_diamond": 26.2, "gsm8k": 1.9, "hellaswag": 63.6, "ifeval": 21.8, "math_500": 1.7, "mmlu_abstract_algebra": 29.0, "mmlu_anatomy": 32.6, "mmlu_astronomy": 20.4, "mmlu_business_ethics": 23.0, "mmlu_clinical_knowledge": 19.6, "mmlu_college_biology": 25.7, "mmlu_college_chemistry": 20.0, "mmlu_college_computer_science": 36.0, "mmlu_college_mathematics": 26.0, "mmlu_college_medicine": 27.2, "mmlu_college_physics": 22.5, "mmlu_computer_security": 27.0, "mmlu_conceptual_physics": 23.0, "mmlu_econometrics": 23.7, "mmlu_electrical_engineering": 30.3, "mmlu_elementary_mathematics": 24.6, "mmlu_formal_logic": 20.6, "mmlu_global_facts": 34.0, "mmlu_high_school_biology": 29.7, "mmlu_high_school_chemistry": 28.6, "mmlu_high_school_computer_science": 31.0, "mmlu_high_school_european_history": 24.8, "mmlu_high_school_geography": 18.2, "mmlu_high_school_government_and_politics": 25.9, "mmlu_high_school_macroeconomics": 27.2, "mmlu_high_school_mathematics": 25.9, "mmlu_high_school_microeconomics": 21.4, "mmlu_high_school_physics": 31.1, "mmlu_high_school_psychology": 24.0, "mmlu_high_school_statistics": 36.6, "mmlu_high_school_us_history": 27.9, "mmlu_high_school_world_history": 28.3, "mmlu_human_aging": 21.5, "mmlu_human_sexuality": 21.4, "mmlu_international_law": 28.1, "mmlu_jurisprudence": 24.1, "mmlu_logical_fallacies": 27.0, "mmlu_machine_learning": 33.0, "mmlu_management": 22.3, "mmlu_marketing": 25.2, "mmlu_medical_genetics": 27.0, "mmlu_miscellaneous": 29.0, "mmlu_moral_disputes": 25.1, "mmlu_moral_scenarios": 24.7, "mmlu_nutrition": 29.7, "mmlu_philosophy": 28.9, "mmlu_prehistory": 25.6, "mmlu_pro": 11.7, "mmlu_professional_accounting": 22.0, "mmlu_professional_law": 24.1, "mmlu_professional_medicine": 40.4, "mmlu_professional_psychology": 27.5, "mmlu_public_relations": 24.5, "mmlu_security_studies": 21.2, "mmlu_sociology": 22.4, "mmlu_us_foreign_policy": 30.0, "mmlu_virology": 22.9, "mmlu_world_religions": 25.7, "musr": 41.0, "truthfulqa": 32.9, "winogrande": 61.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "OLMo 1B hf", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "allen-ai/olmo-1b-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "OLMo 2 0425 1B Instruct", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "allen-ai/olmo-2-0425-1b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "OLMo 2 0425 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "allen-ai/olmo-2-0425-1b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 40.2, "gpqa_diamond": 27.9, "ifeval": 72.4, "math_500": 14.9, "mmlu_pro": 26.7, "musr": 35.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "OLMo 2 1124 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmo-2-1124-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Olmo 3.1 32B Instruct", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "allen-ai/olmo-3-1-32b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "Olmo 3 1025 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmo-3-1025-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "Olmo 3 1125 32B", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "allen-ai/olmo-3-1125-32b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "Olmo 3 7B Instruct SFT", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmo-3-7b-instruct-sft", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "Olmo 3 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmo-3-7b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "Olmo 3 7B Think", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmo-3-7b-think", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Olmo Hybrid 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmo-hybrid-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "olmOCR 2 7B 1025 FP8", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmocr-2-7b-1025-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "olmOCR 2 7B 1025", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmocr-2-7b-1025", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "olmOCR 7B 0225 preview", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/olmocr-7b-0225-preview", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 38.2, "gpqa_diamond": 26.0, "ifeval": 67.6, "math_500": 9.0, "mmlu_pro": 19.1, "musr": 36.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "OLMoE 1B 7B 0125 Instruct", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "allen-ai/olmoe-1b-7b-0125-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 38.2, "gpqa_diamond": 26.0, "ifeval": 67.6, "math_500": 9.0, "mmlu_pro": 19.1, "musr": 36.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "OLMoE 1B 7B 0125", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "allen-ai/olmoe-1b-7b-0125", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 33.9, "gpqa_diamond": 24.7, "ifeval": 21.8, "math_500": 1.7, "mmlu_pro": 17.4, "musr": 34.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "OLMoE 1B 7B 0924", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "allen-ai/olmoe-1b-7b-0924", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "scibert scivocab uncased", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "allen-ai/scibert-scivocab-uncased", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "specter2 base", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "allen-ai/specter2-base", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "unifiedqa t5 small", "fits": {"amd_rx_7900_xt": 18666.7, "amd_rx_7900_xtx": 22400.0, "amd_rx_9070_xt": 15050.0, "apple_m1_max": 9333.3, "apple_m2_max": 9333.3, "apple_m2_ultra": 18666.7, "apple_m3_max": 9333.3, "apple_m3_ultra": 19110.0, "apple_m4": 2800.0, "apple_m4_max": 12740.0, "apple_m4_pro": 6370.0, "nvidia_rtx_3060_12gb": 8400.0, "nvidia_rtx_3090": 21840.0, "nvidia_rtx_4060_ti_16gb": 6720.0, "nvidia_rtx_4070_ti_super": 15680.0, "nvidia_rtx_4080_super": 17173.3, "nvidia_rtx_4090": 23520.0, "nvidia_rtx_5080": 22400.0, "nvidia_rtx_5090": 41813.3}, "model_id": "allen-ai/unifiedqa-t5-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "wildguard", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "allen-ai/wildguard", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Allen AI", "scores_as_of": null}, {"benchmark_scores": {"aider_polyglot": 28.0, "arena_elo_coding": 1270.0, "arena_elo_hard_prompts": 1343.3, "arena_elo_math": 1250.0, "arena_elo_overall": 1260.0, "arena_elo_style_control": 1322.4, "arena_elo_vision": 1126.8, "bbq": 84.5, "gpqa_diamond": 51.2, "helm_safety": 89.8, "humaneval": 82.5, "ifeval": 80.5, "live_code_bench": 28.8, "math_500": 69.5, "medqa": 77.8, "mgsm": 85.6, "mmlu_pro": 65.3, "multipl_e_csharp": 72.5, "multipl_e_julia": 51.8, "multipl_e_kotlin": 62.8, "multipl_e_lua": 46.8, "multipl_e_perl": 42.5, "multipl_e_php": 68.8, "multipl_e_r": 48.5, "multipl_e_ruby": 58.2, "multipl_e_scala": 52.5, "multipl_e_swift": 56.5, "swe_bench_verified": 40.6, "toxigen": 92.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 0.8, "display_name": "Claude Haiku 3.5", "fits": {}, "model_id": "anthropic/claude-3-5-haiku-20241022", "model_subtypes": ["llm-code"], "model_type": "vlm", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 28.0, "arena_elo_coding": 1270.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1260.0, "bbq": 84.5, "gpqa_diamond": 51.2, "helm_safety": 89.8, "humaneval": 82.5, "ifeval": 80.5, "live_code_bench": 28.8, "math_500": 69.5, "medqa": 77.8, "mgsm": 85.6, "mmlu_pro": 65.3, "multipl_e_csharp": 72.5, "multipl_e_julia": 51.8, "multipl_e_kotlin": 62.8, "multipl_e_lua": 46.8, "multipl_e_perl": 42.5, "multipl_e_php": 68.8, "multipl_e_r": 48.5, "multipl_e_ruby": 58.2, "multipl_e_scala": 52.5, "multipl_e_swift": 56.5, "swe_bench_verified": 40.6, "toxigen": 92.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 0.8, "display_name": "Claude Haiku 3.5 (latest)", "fits": {}, "model_id": "anthropic/claude-3-5-haiku-latest", "model_subtypes": ["llm-code"], "model_type": "vlm", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 80.2, "aider_polyglot": 51.6, "alpaca_eval": 52.4, "arena_elo_coding": 1310.0, "arena_elo_hard_prompts": 1358.7, "arena_elo_math": 1290.0, "arena_elo_overall": 1300.0, "arena_elo_style_control": 1341.3, "arena_elo_vision": 1145.5, "bbq": 87.2, "chartqa": 90.8, "docvqa": 95.2, "flores_en_de": 66.9, "flores_en_es": 70.8, "flores_en_ja": 60.8, "flores_en_zh": 59.2, "gpqa_diamond": 65.0, "gsm8k": 96.4, "helm_safety": 91.5, "humaneval": 86.8, "ifeval": 85.4, "live_code_bench": 38.1, "math_500": 78.3, "mathvista": 67.7, "mgsm": 91.6, "mmlu_pro": 76.2, "mmmu": 65.9, "multipl_e_csharp": 80.2, "multipl_e_julia": 61.5, "multipl_e_kotlin": 72.8, "multipl_e_lua": 55.8, "multipl_e_perl": 52.2, "multipl_e_php": 77.5, "multipl_e_r": 58.2, "multipl_e_ruby": 68.5, "multipl_e_scala": 62.5, "multipl_e_swift": 66.5, "ocrbench": 78.8, "swe_bench_verified": 49.0, "toxigen": 94.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 3.5", "fits": {}, "model_id": "anthropic/claude-3-5-sonnet-20240620", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 80.2, "aider_polyglot": 51.6, "arena_elo_coding": 1310.0, "arena_elo_hard_prompts": 1396.7, "arena_elo_math": 1290.0, "arena_elo_overall": 1300.0, "arena_elo_style_control": 1371.4, "arena_elo_vision": 1160.4, "bbq": 87.2, "chartqa": 90.8, "docvqa": 95.2, "flores_en_de": 66.9, "flores_en_es": 70.8, "flores_en_ja": 60.8, "flores_en_zh": 59.2, "gpqa_diamond": 65.0, "gsm8k": 96.4, "helm_safety": 91.5, "humaneval": 86.8, "ifeval": 85.4, "live_code_bench": 38.1, "math_500": 78.3, "mathvista": 67.7, "mgsm": 91.6, "mmlu_pro": 76.2, "mmmu": 65.9, "multipl_e_csharp": 80.2, "multipl_e_julia": 61.5, "multipl_e_kotlin": 72.8, "multipl_e_lua": 55.8, "multipl_e_perl": 52.2, "multipl_e_php": 77.5, "multipl_e_r": 58.2, "multipl_e_ruby": 68.5, "multipl_e_scala": 62.5, "multipl_e_swift": 66.5, "ocrbench": 78.8, "swe_bench_verified": 49.0, "toxigen": 94.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 3.5 v2", "fits": {}, "model_id": "anthropic/claude-3-5-sonnet-20241022", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 64.9, "arena_elo_coding": 1330.0, "arena_elo_hard_prompts": 1415.7, "arena_elo_math": 1310.0, "arena_elo_overall": 1320.0, "arena_elo_style_control": 1386.3, "arena_elo_vision": 1196.1, "gpqa_diamond": 68.0, "humaneval": 88.2, "ifeval": 87.0, "live_code_bench": 47.3, "math_500": 85.5, "medqa": 87.6, "mmlu_pro": 78.0, "multipl_e_csharp": 81.5, "multipl_e_julia": 63.2, "multipl_e_kotlin": 74.5, "multipl_e_lua": 57.8, "multipl_e_perl": 54.5, "multipl_e_php": 78.8, "multipl_e_r": 60.5, "multipl_e_ruby": 70.5, "multipl_e_scala": 64.2, "multipl_e_swift": 68.2, "swe_bench_verified": 49.0, "terminal_bench": 35.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 3.7", "fits": {}, "model_id": "anthropic/claude-3-7-sonnet-20250219", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 88.7, "arena_elo_coding": 1170.0, "arena_elo_hard_prompts": 1263.2, "arena_elo_math": 1160.0, "arena_elo_overall": 1180.0, "arena_elo_style_control": 1259.9, "arena_elo_vision": 1000.8, "bbq": 83.5, "chartqa": 81.1, "docvqa": 88.8, "gpqa_diamond": 41.2, "gsm8k": 88.9, "helm_safety": 88.2, "humaneval": 75.9, "ifeval": 72.0, "math_500": 50.3, "mathvista": 47.9, "mgsm": 75.1, "mmlu_pro": 55.8, "mmmu": 50.2, "toxigen": 91.5}, "capability_tiers": {"coding": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 0.25, "display_name": "Claude Haiku 3", "fits": {}, "model_id": "anthropic/claude-3-haiku-20240307", "model_subtypes": ["llm-code"], "model_type": "vlm", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 88.1, "alpaca_eval": 40.5, "arena_elo_coding": 1250.0, "arena_elo_hard_prompts": 1327.0, "arena_elo_math": 1240.0, "arena_elo_overall": 1260.0, "arena_elo_style_control": 1320.7, "arena_elo_vision": 1062.6, "bbq": 86.5, "chartqa": 80.8, "docvqa": 89.3, "gpqa_diamond": 59.4, "gsm8k": 95.0, "helm_safety": 90.8, "humaneval": 84.9, "ifeval": 81.0, "math_500": 60.1, "mathvista": 50.5, "mgsm": 90.7, "mmlu_pro": 68.5, "mmmu": 59.4, "multipl_e_csharp": 75.5, "multipl_e_julia": 55.8, "multipl_e_kotlin": 67.5, "multipl_e_lua": 50.5, "multipl_e_perl": 46.2, "multipl_e_php": 72.8, "multipl_e_r": 52.5, "multipl_e_ruby": 62.5, "multipl_e_scala": 56.2, "multipl_e_swift": 60.5, "toxigen": 93.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 15.0, "display_name": "Claude Opus 3", "fits": {}, "model_id": "anthropic/claude-3-opus-20240229", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 88.1, "alpaca_eval": 34.9, "arena_elo_coding": 1190.0, "arena_elo_hard_prompts": 1280.5, "arena_elo_math": 1180.0, "arena_elo_overall": 1200.0, "arena_elo_style_control": 1279.8, "arena_elo_vision": 1017.7, "bbq": 85.2, "chartqa": 80.8, "docvqa": 89.3, "gpqa_diamond": 45.5, "gsm8k": 92.3, "helm_safety": 89.5, "humaneval": 73.0, "ifeval": 75.5, "math_500": 55.7, "mathvista": 50.5, "mgsm": 83.5, "mmlu_pro": 60.1, "mmmu": 53.1, "toxigen": 92.8}, "capability_tiers": {"coding": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 3", "fits": {}, "model_id": "anthropic/claude-3-sonnet-20240229", "model_subtypes": ["llm-code"], "model_type": "vlm", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1320.0, "arena_elo_hard_prompts": 1436.6, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "arena_elo_style_control": 1407.2, "gpqa_diamond": 62.1, "humaneval": 86.2, "ifeval": 85.0, "live_code_bench": 51.1, "math_500": 84.3, "medqa": 77.8, "mmlu_pro": 75.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 72.2, "multipl_e_lua": 54.8, "multipl_e_perl": 50.5, "multipl_e_php": 75.8, "multipl_e_r": 56.8, "multipl_e_ruby": 66.2, "multipl_e_scala": 60.5, "multipl_e_swift": 65.5, "swe_bench_verified": 73.3, "terminal_bench": 41.0}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 1.0, "display_name": "Claude Haiku 4.5", "fits": {}, "model_id": "anthropic/claude-haiku-4-5-20251001", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1320.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "gpqa_diamond": 62.1, "humaneval": 86.2, "ifeval": 85.0, "live_code_bench": 51.1, "math_500": 84.3, "medqa": 77.8, "mmlu_pro": 75.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 72.2, "multipl_e_lua": 54.8, "multipl_e_perl": 50.5, "multipl_e_php": 75.8, "multipl_e_r": 56.8, "multipl_e_ruby": 66.2, "multipl_e_scala": 60.5, "multipl_e_swift": 65.5, "swe_bench_verified": 73.3, "terminal_bench": 41.0}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 1.0, "display_name": "Claude Haiku 4.5 (latest)", "fits": {}, "model_id": "anthropic/claude-haiku-4-5", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_overall": 1410.0, "browsecomp": 86.9, "charxiv_reasoning": 86.1, "charxiv_reasoning_tools": 93.2, "gpqa_diamond": 94.55, "graphwalks_bfs_256k_1m": 80.0, "graphwalks_parents_256k_1m": 97.7, "hle": 56.8, "hle_tools": 64.7, "humaneval": 93.2, "ifeval": 92.1, "lab_bench_figqa": 79.7, "lab_bench_figqa_tools": 89.0, "math_500": 96.4, "mmlu_pro": 85.2, "mmmlu": 92.67, "mt_bench": 9.4, "osworld": 79.6, "screenspot_pro": 79.5, "screenspot_pro_tools": 92.8, "swe_bench_multilingual": 87.3, "swe_bench_multimodal": 59.0, "swe_bench_pro": 77.8, "swe_bench_verified": 93.9, "terminal_bench_2": 82.0, "usamo_2026": 97.6}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:computer_use": "tier-1", "tool_use:function_calling": "tier-1", "tool_use:mcp_compatible": "tier-1", "tool_use:multi_turn_tool_use": "tier-1", "tool_use:parallel_tool_calls": "tier-1", "tool_use:tool_error_recovery": "tier-1"}, "context_window": 1000000, "cost_input": 15.0, "display_name": "Claude Mythos Preview", "fits": {}, "model_id": "anthropic/claude-mythos-preview", "model_subtypes": ["llm-code", "vlm"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.0, "arena_elo_coding": 1380.0, "arena_elo_math": 1360.0, "arena_elo_overall": 1370.0, "gpqa_diamond": 73.8, "humaneval": 90.5, "ifeval": 89.7, "live_code_bench": 54.2, "math_500": 92.8, "mmlu_pro": 82.1, "multipl_e_csharp": 84.5, "multipl_e_julia": 66.5, "multipl_e_kotlin": 78.2, "multipl_e_lua": 61.2, "multipl_e_perl": 57.8, "multipl_e_php": 81.8, "multipl_e_r": 64.2, "multipl_e_ruby": 74.5, "multipl_e_scala": 67.8, "multipl_e_swift": 72.1, "swe_bench_verified": 62.3, "terminal_bench": 39.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 15.0, "display_name": "Claude Opus 4 (latest)", "fits": {}, "model_id": "anthropic/claude-opus-4-0", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 78.8, "arena_elo_coding": 1400.0, "arena_elo_hard_prompts": 1479.9, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "arena_elo_style_control": 1448.6, "gpqa_diamond": 76.2, "humaneval": 91.7, "ifeval": 91.0, "live_code_bench": 65.4, "math_500": 94.5, "mmlu_pro": 83.8, "multipl_e_csharp": 86.2, "multipl_e_julia": 68.2, "multipl_e_kotlin": 80.5, "multipl_e_lua": 63.5, "multipl_e_perl": 60.2, "multipl_e_php": 83.5, "multipl_e_r": 66.5, "multipl_e_ruby": 76.2, "multipl_e_scala": 70.1, "multipl_e_swift": 74.5, "swe_bench_agent": 58.2, "swe_bench_verified": 68.8, "tau_bench": 64.8, "terminal_bench": 52.1}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 15.0, "display_name": "Claude Opus 4.1", "fits": {}, "model_id": "anthropic/claude-opus-4-1-20250805", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 78.8, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 76.2, "humaneval": 91.7, "ifeval": 91.0, "live_code_bench": 65.4, "math_500": 94.5, "mmlu_pro": 83.8, "multipl_e_csharp": 86.2, "multipl_e_julia": 68.2, "multipl_e_kotlin": 80.5, "multipl_e_lua": 63.5, "multipl_e_perl": 60.2, "multipl_e_php": 83.5, "multipl_e_r": 66.5, "multipl_e_ruby": 76.2, "multipl_e_scala": 70.1, "multipl_e_swift": 74.5, "swe_bench_agent": 58.2, "swe_bench_verified": 68.8, "tau_bench": 64.8, "terminal_bench": 52.1}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 15.0, "display_name": "Claude Opus 4.1 (latest)", "fits": {}, "model_id": "anthropic/claude-opus-4-1", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 82.1, "aime_2025": 82.0, "alpaca_eval": 55.2, "arena_elo_coding": 1420.0, "arena_elo_hard_prompts": 1455.5, "arena_elo_math": 1400.0, "arena_elo_overall": 1410.0, "arena_elo_style_control": 1423.7, "arena_elo_vision": 1207.5, "bbq": 88.2, "chartqa": 86.8, "docvqa": 93.2, "finbench": 71.2, "gpqa_diamond": 78.4, "helm_safety": 92.5, "humaneval": 93.2, "ifeval": 92.1, "legalbench": 75.5, "live_code_bench": 62.4, "math_500": 96.4, "mathvista": 65.5, "medqa": 82.1, "mmlu_astronomy": 80.5, "mmlu_biology": 88.2, "mmlu_business_ethics": 82.1, "mmlu_chemistry": 82.5, "mmlu_clinical_knowledge": 87.5, "mmlu_computer_science": 88.8, "mmlu_jurisprudence": 80.5, "mmlu_physics": 85.1, "mmlu_pro": 85.2, "mmlu_professional_accounting": 72.5, "mmlu_professional_law": 78.2, "mmmu": 70.8, "mt_bench": 9.4, "multipl_e": 89.3, "multipl_e_cpp": 88.5, "multipl_e_csharp": 88.2, "multipl_e_go": 85.5, "multipl_e_java": 91.2, "multipl_e_javascript": 91.5, "multipl_e_julia": 70.2, "multipl_e_kotlin": 82.5, "multipl_e_lua": 65.8, "multipl_e_perl": 62.5, "multipl_e_php": 85.5, "multipl_e_python": 95.5, "multipl_e_r": 68.5, "multipl_e_ruby": 78.5, "multipl_e_rust": 82.1, "multipl_e_scala": 72.2, "multipl_e_swift": 76.8, "multipl_e_typescript": 90.8, "swe_bench_agent": 62.5, "swe_bench_verified": 72.7, "tau_bench": 68.2, "terminal_bench": 55.8, "terminal_bench_2": 65.4, "toxigen": 95.1, "wildbench": 82.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 15.0, "display_name": "Claude Opus 4", "fits": {}, "model_id": "anthropic/claude-opus-4-20250514", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aime_2025": 78.0, "arena_elo_coding": 1410.0, "arena_elo_hard_prompts": 1500.0, "arena_elo_math": 1390.0, "arena_elo_overall": 1400.0, "arena_elo_style_control": 1473.1, "gpqa_diamond": 77.0, "humaneval": 92.5, "ifeval": 91.5, "live_code_bench": 73.8, "math_500": 95.2, "mmlu_pro": 84.5, "multipl_e_csharp": 87.5, "multipl_e_julia": 69.5, "multipl_e_kotlin": 81.8, "multipl_e_lua": 64.5, "multipl_e_perl": 61.2, "multipl_e_php": 84.8, "multipl_e_r": 67.8, "multipl_e_ruby": 77.2, "multipl_e_scala": 71.5, "multipl_e_swift": 75.5, "swe_bench_verified": 70.3, "terminal_bench_2": 59.3}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 5.0, "display_name": "Claude Opus 4.5", "fits": {}, "model_id": "anthropic/claude-opus-4-5-20251101", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aime_2025": 78.0, "arena_elo_coding": 1410.0, "arena_elo_math": 1390.0, "arena_elo_overall": 1400.0, "gpqa_diamond": 77.0, "humaneval": 92.5, "ifeval": 91.5, "live_code_bench": 73.8, "math_500": 95.2, "mmlu_pro": 84.5, "multipl_e_csharp": 87.5, "multipl_e_julia": 69.5, "multipl_e_kotlin": 81.8, "multipl_e_lua": 64.5, "multipl_e_perl": 61.2, "multipl_e_php": 84.8, "multipl_e_r": 67.8, "multipl_e_ruby": 77.2, "multipl_e_scala": 71.5, "multipl_e_swift": 75.5, "swe_bench_verified": 70.3, "terminal_bench_2": 59.3}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 5.0, "display_name": "Claude Opus 4.5 (latest)", "fits": {}, "model_id": "anthropic/claude-opus-4-5", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 82.1, "aime_2025": 82.0, "alpaca_eval": 55.2, "arena_elo_coding": 1420.0, "arena_elo_hard_prompts": 1534.7, "arena_elo_math": 1400.0, "arena_elo_overall": 1410.0, "arena_elo_style_control": 1502.8, "arena_elo_vision": 1295.0, "artificial_analysis_quality_index": 88.0, "artificial_analysis_speed_index": 62.0, "bbq": 88.2, "browsecomp": 83.7, "chartqa": 86.8, "charxiv_reasoning": 61.5, "charxiv_reasoning_tools": 78.9, "docvqa": 93.2, "finbench": 71.2, "gpqa_diamond": 91.3, "graphwalks_bfs_256k_1m": 38.7, "helm_safety": 92.5, "hle": 40.0, "hle_tools": 53.1, "humaneval": 93.2, "ifeval": 92.1, "lab_bench_figqa": 58.5, "lab_bench_figqa_tools": 75.1, "legalbench": 75.5, "live_code_bench": 62.4, "math_500": 96.4, "mathvista": 65.5, "medqa": 82.1, "mmlu_astronomy": 80.5, "mmlu_biology": 88.2, "mmlu_business_ethics": 82.1, "mmlu_chemistry": 82.5, "mmlu_clinical_knowledge": 87.5, "mmlu_computer_science": 88.8, "mmlu_jurisprudence": 80.5, "mmlu_physics": 85.1, "mmlu_pro": 85.2, "mmlu_professional_accounting": 72.5, "mmlu_professional_law": 78.2, "mmmlu": 91.1, "mmmu": 70.8, "mt_bench": 9.4, "multipl_e": 89.3, "multipl_e_cpp": 88.5, "multipl_e_csharp": 88.2, "multipl_e_go": 85.5, "multipl_e_java": 91.2, "multipl_e_javascript": 91.5, "multipl_e_julia": 70.2, "multipl_e_kotlin": 82.5, "multipl_e_lua": 65.8, "multipl_e_perl": 62.5, "multipl_e_php": 85.5, "multipl_e_python": 95.5, "multipl_e_r": 68.5, "multipl_e_ruby": 78.5, "multipl_e_rust": 82.1, "multipl_e_scala": 72.2, "multipl_e_swift": 76.8, "multipl_e_typescript": 90.8, "osworld": 72.7, "screenspot_pro": 57.7, "screenspot_pro_tools": 83.1, "swe_bench_agent": 62.5, "swe_bench_multilingual": 77.8, "swe_bench_multimodal": 27.1, "swe_bench_pro": 53.4, "swe_bench_verified": 80.8, "tau_bench": 68.2, "terminal_bench": 55.8, "terminal_bench_2": 65.4, "toxigen": 95.1, "usamo_2026": 42.3, "wildbench": 82.5}, "capability_tiers": {"coding": "tier-1", "coding:agentic_coding": "tier-1", "coding:code_completion": "tier-1", "coding:code_review": "tier-1", "coding:debugging": "tier-1", "coding:documentation": "tier-1", "coding:multi_file_editing": "tier-1", "coding:refactoring": "tier-1", "coding:repository_understanding": "tier-1", "coding:test_generation": "tier-1", "reasoning": "tier-1", "reasoning:causal": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:logical": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "reasoning:planning": "tier-1", "reasoning:scientific": "tier-1", "reasoning:self_correction": "tier-1", "reasoning:think_budget_control": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1", "tool_use:mcp_compatible": "tier-1", "tool_use:parallel_tool_calls": "tier-1"}, "context_window": 1000000, "cost_input": 5.0, "display_name": "Claude Opus 4.6", "fits": {}, "model_id": "anthropic/claude-opus-4-6", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 61.3, "arena_elo_coding": 1350.0, "arena_elo_math": 1330.0, "arena_elo_overall": 1340.0, "artificial_analysis_quality_index": 85.0, "artificial_analysis_speed_index": 73.0, "gpqa_diamond": 68.4, "humaneval": 88.7, "ifeval": 87.2, "live_code_bench": 65.5, "math_500": 88.6, "mmlu_pro": 78.5, "multipl_e_csharp": 82.5, "multipl_e_julia": 64.5, "multipl_e_kotlin": 76.2, "multipl_e_lua": 58.8, "multipl_e_perl": 55.5, "multipl_e_php": 79.8, "multipl_e_r": 62.1, "multipl_e_ruby": 72.5, "multipl_e_scala": 65.8, "multipl_e_swift": 70.1, "swe_bench_verified": 55.2, "terminal_bench": 35.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 4 (latest)", "fits": {}, "model_id": "anthropic/claude-sonnet-4-0", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 75.5, "alpaca_eval": 50.8, "arena_elo_coding": 1390.0, "arena_elo_hard_prompts": 1430.5, "arena_elo_math": 1370.0, "arena_elo_overall": 1380.0, "arena_elo_style_control": 1398.5, "arena_elo_vision": 1207.0, "artificial_analysis_quality_index": 85.0, "artificial_analysis_speed_index": 73.0, "bbq": 87.5, "chartqa": 84.2, "docvqa": 91.5, "gpqa_diamond": 74.1, "helm_safety": 91.8, "humaneval": 91.5, "ifeval": 90.4, "live_code_bench": 59.0, "math_500": 93.8, "mathvista": 62.1, "mmlu_astronomy": 77.8, "mmlu_biology": 85.5, "mmlu_business_ethics": 79.5, "mmlu_chemistry": 79.8, "mmlu_clinical_knowledge": 84.2, "mmlu_computer_science": 86.2, "mmlu_jurisprudence": 77.8, "mmlu_physics": 82.5, "mmlu_pro": 83.0, "mmlu_professional_accounting": 69.8, "mmlu_professional_law": 75.5, "mmmu": 68.2, "mt_bench": 9.2, "multipl_e": 86.7, "multipl_e_cpp": 85.8, "multipl_e_csharp": 85.5, "multipl_e_go": 82.8, "multipl_e_java": 89.1, "multipl_e_javascript": 89.5, "multipl_e_julia": 67.5, "multipl_e_kotlin": 79.8, "multipl_e_lua": 62.1, "multipl_e_perl": 58.8, "multipl_e_php": 82.2, "multipl_e_python": 93.2, "multipl_e_r": 65.2, "multipl_e_ruby": 75.2, "multipl_e_rust": 78.5, "multipl_e_scala": 68.5, "multipl_e_swift": 73.5, "multipl_e_typescript": 88.2, "swe_bench_agent": 55.8, "swe_bench_verified": 65.3, "tau_bench": 61.5, "terminal_bench": 48.2, "terminal_bench_2": 59.1, "toxigen": 94.5, "wildbench": 78.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 4", "fits": {}, "model_id": "anthropic/claude-sonnet-4-20250514", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 75.5, "alpaca_eval": 50.8, "arena_elo_coding": 1380.0, "arena_elo_hard_prompts": 1485.0, "arena_elo_math": 1360.0, "arena_elo_overall": 1370.0, "arena_elo_style_control": 1451.9, "artificial_analysis_quality_index": 86.0, "artificial_analysis_speed_index": 71.0, "bbq": 87.5, "chartqa": 84.2, "docvqa": 91.5, "gpqa_diamond": 72.5, "helm_safety": 91.8, "humaneval": 90.8, "ifeval": 89.3, "live_code_bench": 59.0, "math_500": 92.1, "mathvista": 62.1, "mmlu_astronomy": 77.8, "mmlu_biology": 85.5, "mmlu_business_ethics": 79.5, "mmlu_chemistry": 79.8, "mmlu_clinical_knowledge": 84.2, "mmlu_computer_science": 86.2, "mmlu_jurisprudence": 77.8, "mmlu_physics": 82.5, "mmlu_pro": 81.7, "mmlu_professional_accounting": 69.8, "mmlu_professional_law": 75.5, "mmmu": 68.2, "mt_bench": 9.2, "multipl_e": 86.7, "multipl_e_cpp": 85.8, "multipl_e_csharp": 85.5, "multipl_e_go": 82.8, "multipl_e_java": 89.1, "multipl_e_javascript": 89.5, "multipl_e_julia": 67.5, "multipl_e_kotlin": 79.8, "multipl_e_lua": 62.1, "multipl_e_perl": 58.8, "multipl_e_php": 82.2, "multipl_e_python": 93.2, "multipl_e_r": 65.2, "multipl_e_ruby": 75.2, "multipl_e_rust": 78.5, "multipl_e_scala": 68.5, "multipl_e_swift": 73.5, "multipl_e_typescript": 88.2, "swe_bench_agent": 55.8, "swe_bench_verified": 62.1, "tau_bench": 61.5, "terminal_bench": 48.2, "toxigen": 94.5, "wildbench": 78.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 4.5", "fits": {}, "model_id": "anthropic/claude-sonnet-4-5-20250929", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 75.5, "alpaca_eval": 50.8, "arena_elo_coding": 1380.0, "arena_elo_math": 1360.0, "arena_elo_overall": 1370.0, "artificial_analysis_quality_index": 86.0, "artificial_analysis_speed_index": 71.0, "bbq": 87.5, "chartqa": 84.2, "docvqa": 91.5, "gpqa_diamond": 72.5, "helm_safety": 91.8, "humaneval": 90.8, "ifeval": 89.3, "live_code_bench": 59.0, "math_500": 92.1, "mathvista": 62.1, "mmlu_astronomy": 77.8, "mmlu_biology": 85.5, "mmlu_business_ethics": 79.5, "mmlu_chemistry": 79.8, "mmlu_clinical_knowledge": 84.2, "mmlu_computer_science": 86.2, "mmlu_jurisprudence": 77.8, "mmlu_physics": 82.5, "mmlu_pro": 81.7, "mmlu_professional_accounting": 69.8, "mmlu_professional_law": 75.5, "mmmu": 68.2, "mt_bench": 9.2, "multipl_e": 86.7, "multipl_e_cpp": 85.8, "multipl_e_csharp": 85.5, "multipl_e_go": 82.8, "multipl_e_java": 89.1, "multipl_e_javascript": 89.5, "multipl_e_julia": 67.5, "multipl_e_kotlin": 79.8, "multipl_e_lua": 62.1, "multipl_e_perl": 58.8, "multipl_e_php": 82.2, "multipl_e_python": 93.2, "multipl_e_r": 65.2, "multipl_e_ruby": 75.2, "multipl_e_rust": 78.5, "multipl_e_scala": 68.5, "multipl_e_swift": 73.5, "multipl_e_typescript": 88.2, "swe_bench_agent": 55.8, "swe_bench_verified": 62.1, "tau_bench": 61.5, "terminal_bench": 48.2, "toxigen": 94.5, "wildbench": 78.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 3.0, "display_name": "Claude Sonnet 4.5 (latest)", "fits": {}, "model_id": "anthropic/claude-sonnet-4-5", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1390.0, "arena_elo_hard_prompts": 1498.7, "arena_elo_math": 1370.0, "arena_elo_overall": 1380.0, "arena_elo_style_control": 1462.2, "arena_elo_vision": 1271.1, "artificial_analysis_quality_index": 85.0, "artificial_analysis_speed_index": 73.0, "gpqa_diamond": 74.1, "humaneval": 91.5, "ifeval": 90.4, "math_500": 93.8, "mmlu_pro": 83.0, "multipl_e_csharp": 86.8, "multipl_e_julia": 68.5, "multipl_e_kotlin": 80.8, "multipl_e_lua": 63.8, "multipl_e_perl": 60.5, "multipl_e_php": 83.5, "multipl_e_r": 66.8, "multipl_e_ruby": 76.5, "multipl_e_scala": 70.2, "multipl_e_swift": 74.8, "swe_bench_verified": 65.3, "terminal_bench_2": 59.1}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 3.0, "display_name": "Claude Sonnet 4.6", "fits": {}, "model_id": "anthropic/claude-sonnet-4-6", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Anthropic", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "AltCLIP", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "baai/altclip", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "AquilaChat 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baai/aquilachat-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "AquilaChat2 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baai/aquilachat2-7b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {"beir": 50.2, "miracl": 39.0, "mteb_classification": 68.5, "mteb_clustering": 45.8, "mteb_overall": 62.1, "mteb_pair_classification": 83.2, "mteb_reranking": 54.8, "mteb_retrieval": 57.8, "mteb_sts": 80.5, "mteb_summarization": 28.8}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "bge base en v1.5", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "baai/bge-base-en-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 49.0, "miracl": 37.5, "mteb_classification": 67.2, "mteb_clustering": 44.5, "mteb_overall": 60.8, "mteb_pair_classification": 82.5, "mteb_reranking": 53.8, "mteb_retrieval": 56.5, "mteb_sts": 79.5, "mteb_summarization": 28.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "bge base en", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "baai/bge-base-en", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 68.2, "mteb_clustering": 45.5, "mteb_overall": 62.0, "mteb_retrieval": 57.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "bge base zh v1.5", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "baai/bge-base-zh-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 67.5, "mteb_clustering": 44.8, "mteb_overall": 61.0, "mteb_retrieval": 56.8}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "bge base zh", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "baai/bge-base-zh", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 66.8, "mteb_clustering": 43.5, "mteb_overall": 60.5, "mteb_pair_classification": 82.0, "mteb_reranking": 53.0, "mteb_retrieval": 57.2, "mteb_sts": 78.5, "mteb_summarization": 27.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "bge code v1", "fits": {"amd_rx_7900_xt": 1971.8, "amd_rx_7900_xtx": 2366.2, "amd_rx_9070_xt": 1589.8, "apple_m1_max": 985.9, "apple_m2_max": 985.9, "apple_m2_ultra": 1971.8, "apple_m3_max": 985.9, "apple_m3_ultra": 2018.7, "apple_m4": 295.8, "apple_m4_max": 1345.8, "apple_m4_pro": 672.9, "nvidia_rtx_3060_12gb": 887.3, "nvidia_rtx_3090": 2307.0, "nvidia_rtx_4060_ti_16gb": 709.9, "nvidia_rtx_4070_ti_super": 1656.3, "nvidia_rtx_4080_super": 1814.1, "nvidia_rtx_4090": 2484.5, "nvidia_rtx_5080": 2366.2, "nvidia_rtx_5090": 4416.9}, "model_id": "baai/bge-code-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 52.1, "miracl": 42.5, "mteb_classification": 70.3, "mteb_clustering": 48.2, "mteb_overall": 64.2, "mteb_pair_classification": 85.0, "mteb_reranking": 56.5, "mteb_retrieval": 60.5, "mteb_sts": 82.5, "mteb_summarization": 30.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "bge large en v1.5", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "baai/bge-large-en-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 51.0, "miracl": 41.0, "mteb_classification": 69.1, "mteb_clustering": 47.0, "mteb_overall": 63.0, "mteb_pair_classification": 84.0, "mteb_reranking": 55.5, "mteb_retrieval": 59.2, "mteb_sts": 81.5, "mteb_summarization": 29.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "bge large en", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "baai/bge-large-en", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 70.8, "mteb_clustering": 48.5, "mteb_overall": 64.5, "mteb_retrieval": 61.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "bge large zh v1.5", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "baai/bge-large-zh-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 69.8, "mteb_clustering": 47.5, "mteb_overall": 63.5, "mteb_retrieval": 60.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "bge large zh", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "baai/bge-large-zh", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 54.8, "miracl": 67.8, "mteb_classification": 72.1, "mteb_clustering": 52.3, "mteb_overall": 68.2, "mteb_pair_classification": 84.5, "mteb_reranking": 57.5, "mteb_retrieval": 65.8, "mteb_sts": 82.0, "mteb_summarization": 29.8}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "bge m3", "fits": {"amd_rx_7900_xt": 1971.8, "amd_rx_7900_xtx": 2366.2, "amd_rx_9070_xt": 1589.8, "apple_m1_max": 985.9, "apple_m2_max": 985.9, "apple_m2_ultra": 1971.8, "apple_m3_max": 985.9, "apple_m3_ultra": 2018.7, "apple_m4": 295.8, "apple_m4_max": 1345.8, "apple_m4_pro": 672.9, "nvidia_rtx_3060_12gb": 887.3, "nvidia_rtx_3090": 2307.0, "nvidia_rtx_4060_ti_16gb": 709.9, "nvidia_rtx_4070_ti_super": 1656.3, "nvidia_rtx_4080_super": 1814.1, "nvidia_rtx_4090": 2484.5, "nvidia_rtx_5080": 2366.2, "nvidia_rtx_5090": 4416.9}, "model_id": "baai/bge-m3", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 59.0, "mteb_classification": 71.2, "mteb_clustering": 50.5, "mteb_overall": 66.8, "mteb_pair_classification": 85.0, "mteb_reranking": 58.0, "mteb_retrieval": 63.5, "mteb_sts": 83.5, "mteb_summarization": 30.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "bge multilingual gemma2", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "baai/bge-multilingual-gemma2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "bge reranker base", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "baai/bge-reranker-base", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "bge reranker large", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "baai/bge-reranker-large", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "bge reranker v2 gemma", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "baai/bge-reranker-v2-gemma", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {"mteb_overall": 70.1, "mteb_retrieval": 68.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "bge reranker v2 m3", "fits": {"amd_rx_7900_xt": 1971.8, "amd_rx_7900_xtx": 2366.2, "amd_rx_9070_xt": 1589.8, "apple_m1_max": 985.9, "apple_m2_max": 985.9, "apple_m2_ultra": 1971.8, "apple_m3_max": 985.9, "apple_m3_ultra": 2018.7, "apple_m4": 295.8, "apple_m4_max": 1345.8, "apple_m4_pro": 672.9, "nvidia_rtx_3060_12gb": 887.3, "nvidia_rtx_3090": 2307.0, "nvidia_rtx_4060_ti_16gb": 709.9, "nvidia_rtx_4070_ti_super": 1656.3, "nvidia_rtx_4080_super": 1814.1, "nvidia_rtx_4090": 2484.5, "nvidia_rtx_5080": 2366.2, "nvidia_rtx_5090": 4416.9}, "model_id": "baai/bge-reranker-v2-m3", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 47.5, "miracl": 35.5, "mteb_classification": 65.8, "mteb_clustering": 42.1, "mteb_overall": 59.3, "mteb_pair_classification": 81.0, "mteb_reranking": 52.5, "mteb_retrieval": 54.2, "mteb_sts": 78.0, "mteb_summarization": 27.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.005, "display_name": "bge small en v1.5", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "baai/bge-small-en-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 45.8, "miracl": 33.5, "mteb_classification": 64.0, "mteb_clustering": 40.5, "mteb_overall": 57.5, "mteb_pair_classification": 80.5, "mteb_reranking": 51.5, "mteb_retrieval": 52.0, "mteb_sts": 77.0, "mteb_summarization": 26.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.005, "display_name": "bge small en", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "baai/bge-small-en", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 65.5, "mteb_clustering": 42.0, "mteb_overall": 59.0, "mteb_retrieval": 54.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.005, "display_name": "bge small zh v1.5", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "baai/bge-small-zh-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 64.8, "mteb_clustering": 41.2, "mteb_overall": 58.2, "mteb_retrieval": 53.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.005, "display_name": "bge small zh", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "baai/bge-small-zh", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Emu3 Chat hf", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "baai/emu3-chat-hf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Emu3 VisionTokenizer", "fits": {"amd_rx_7900_xt": 3733.3, "amd_rx_7900_xtx": 4480.0, "amd_rx_9070_xt": 3010.0, "apple_m1_max": 1866.7, "apple_m2_max": 1866.7, "apple_m2_ultra": 3733.3, "apple_m3_max": 1866.7, "apple_m3_ultra": 3822.0, "apple_m4": 560.0, "apple_m4_max": 2548.0, "apple_m4_pro": 1274.0, "nvidia_rtx_3060_12gb": 1680.0, "nvidia_rtx_3090": 4368.0, "nvidia_rtx_4060_ti_16gb": 1344.0, "nvidia_rtx_4070_ti_super": 3136.0, "nvidia_rtx_4080_super": 3434.7, "nvidia_rtx_4090": 4704.0, "nvidia_rtx_5080": 4480.0, "nvidia_rtx_5090": 8362.7}, "model_id": "baai/emu3-visiontokenizer", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 49.5, "gpqa_diamond": 27.5, "ifeval": 60.5, "math_500": 8.8, "mmlu_pro": 32.5, "musr": 37.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Infinity Instruct 3M 0625 Llama3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "baai/infinity-instruct-3m-0625-llama3-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "BAAI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 55.1, "gpqa_diamond": 35.4, "ifeval": 51.9, "math_500": 16.4, "mmlu_pro": 41.2, "musr": 45.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": 0.15, "display_name": "Infinity Instruct 3M 0625 Yi 1.5 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "baai/infinity-instruct-3m-0625-yi-1-5-9b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "BAAI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 50.8, "gpqa_diamond": 29.3, "ifeval": 61.3, "math_500": 12.8, "mmlu_pro": 32.2, "musr": 35.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Infinity Instruct 7M Gen Llama3 1 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "baai/infinity-instruct-7m-gen-llama3-1-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "BAAI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "llm embedder", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "baai/llm-embedder", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "RoboBrain2.5 8B NV", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "baai/robobrain2-5-8b-nv", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "seggpt vit large", "fits": {"amd_rx_7900_xt": 3294.1, "amd_rx_7900_xtx": 3952.9, "amd_rx_9070_xt": 2655.9, "apple_m1_max": 1647.1, "apple_m2_max": 1647.1, "apple_m2_ultra": 3294.1, "apple_m3_max": 1647.1, "apple_m3_ultra": 3372.4, "apple_m4": 494.1, "apple_m4_max": 2248.2, "apple_m4_pro": 1124.1, "nvidia_rtx_3060_12gb": 1482.4, "nvidia_rtx_3090": 3854.1, "nvidia_rtx_4060_ti_16gb": 1185.9, "nvidia_rtx_4070_ti_super": 2767.1, "nvidia_rtx_4080_super": 3030.6, "nvidia_rtx_4090": 4150.6, "nvidia_rtx_5080": 3952.9, "nvidia_rtx_5090": 7378.8}, "model_id": "baai/seggpt-vit-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "BAAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan 13B Base", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "baichuan/baichuan-13b-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan 13B Chat", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "baichuan/baichuan-13b-chat", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {"gsm8k": 2.8, "winogrande": 66.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baichuan/baichuan-7b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Baichuan", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan Audio Base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baichuan/baichuan-audio-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan Audio Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baichuan/baichuan-audio-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Baichuan M1 14B Instruct", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "baichuan/baichuan-m1-14b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Baichuan M2 32B GPTQ Int4", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "baichuan/baichuan-m2-32b-gptq-int4", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Baichuan M2 32B", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "baichuan/baichuan-m2-32b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Baichuan M3 235B FP8", "fits": {"apple_m2_ultra": 4.8, "apple_m3_ultra": 4.9}, "model_id": "baichuan/baichuan-m3-235b-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Baichuan M3 235B", "fits": {"apple_m2_ultra": 4.8, "apple_m3_ultra": 4.9}, "model_id": "baichuan/baichuan-m3-235b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Baichuan Omni 1d5", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baichuan/baichuan-omni-1d5", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan2 13B Base", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "baichuan/baichuan2-13b-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan2 13B Chat", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "baichuan/baichuan2-13b-chat", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan2 7B Base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baichuan/baichuan2-7b-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Baichuan2 7B Chat", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "baichuan/baichuan2-7b-chat", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Baichuan", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "FLUX.1 Canny dev lora", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-canny-dev-lora", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "FLUX.1 Depth dev lora", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-depth-dev-lora", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.025, "display_name": "FLUX.1 dev", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-dev", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "FLUX.1 Fill dev", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-fill-dev", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.1 Kontext dev", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-kontext-dev", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.1 Krea dev", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-krea-dev", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "FLUX.1 Redux dev", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-redux-dev", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.003, "display_name": "FLUX.1 schnell", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "black-forest-labs/flux-1-schnell", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "FLUX.2 dev NVFP4", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "black-forest-labs/flux-2-dev-nvfp4", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "FLUX.2 dev", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "black-forest-labs/flux-2-dev", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein 4B fp8", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "black-forest-labs/flux-2-klein-4b-fp8", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "black-forest-labs/flux-2-klein-4b", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein 9B fp8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "black-forest-labs/flux-2-klein-9b-fp8", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein 9B kv fp8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "black-forest-labs/flux-2-klein-9b-kv-fp8", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein 9B kv", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "black-forest-labs/flux-2-klein-9b-kv", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "black-forest-labs/flux-2-klein-9b", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein base 4B fp8", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "black-forest-labs/flux-2-klein-base-4b-fp8", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein base 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "black-forest-labs/flux-2-klein-base-4b", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein base 9B fp8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "black-forest-labs/flux-2-klein-base-9b-fp8", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "FLUX.2 klein base 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "black-forest-labs/flux-2-klein-base-9b", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Black Forest Labs", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.25, "display_name": "GPT OSS 120B", "fits": {"apple_m2_max": 4.7, "apple_m2_ultra": 9.3, "apple_m3_max": 4.7, "apple_m3_ultra": 9.6, "apple_m4_max": 6.4}, "model_id": "cerebras/gpt-oss-120b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Cerebras", "scores_as_of": null}, {"benchmark_scores": {"bbh": 47.8, "gpqa_diamond": 31.0, "ifeval": 33.2, "math_500": 5.7, "mmlu_pro": 31.6, "musr": 39.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32000, "cost_input": 0.1, "display_name": "Llama 3.1 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "cerebras/llama3-1-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cerebras", "scores_as_of": "2025-03"}, {"benchmark_scores": {"aider_polyglot": 68.2, "alpaca_eval": 44.5, "mmlu_astronomy": 75.8, "mmlu_biology": 83.8, "mmlu_business_ethics": 76.5, "mmlu_chemistry": 77.2, "mmlu_clinical_knowledge": 82.2, "mmlu_computer_science": 84.5, "mmlu_jurisprudence": 73.8, "mmlu_physics": 81.5, "mmlu_professional_accounting": 67.5, "mmlu_professional_law": 71.8, "mt_bench": 8.9, "swe_bench_agent": 45.5, "swe_bench_verified": 52.1, "tau_bench": 50.2, "wildbench": 73.8}, "capability_tiers": {"coding": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131000, "cost_input": 0.6, "display_name": "Qwen 3 235B Instruct", "fits": {"apple_m2_ultra": 4.8, "apple_m3_ultra": 4.9}, "model_id": "cerebras/qwen-3-235b-a22b-instruct-2507", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cerebras", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 2.25, "display_name": "Z.AI GLM-4.7", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "cerebras/zai-glm-4-7", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cerebras", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": null, "display_name": "Aya Expanse 32B", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "cohere/c4ai-aya-expanse-32b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8000, "cost_input": null, "display_name": "Aya Expanse 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "cohere/c4ai-aya-expanse-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 16000, "cost_input": null, "display_name": "Aya Vision 32B", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "cohere/c4ai-aya-vision-32b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Cohere", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 16000, "cost_input": null, "display_name": "Aya Vision 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "cohere/c4ai-aya-vision-8b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Cohere", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_hard_prompts": 1367.9, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "arena_elo_style_control": 1353.4, "artificial_analysis_quality_index": 78.0, "artificial_analysis_speed_index": 58.0, "gpqa_diamond": 48.2, "humaneval": 78.5, "ifeval": 78.0, "math_500": 72.5, "mmlu_pro": 63.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 2.5, "display_name": "Command A", "fits": {"apple_m2_max": 5.0, "apple_m2_ultra": 10.1, "apple_m3_max": 5.0, "apple_m3_ultra": 10.3, "apple_m4_max": 6.9}, "model_id": "cohere/command-a-03-2025", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 78.0, "artificial_analysis_speed_index": 58.0, "gpqa_diamond": 48.2, "humaneval": 78.5, "ifeval": 78.0, "math_500": 72.5, "mmlu_pro": 63.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 2.5, "display_name": "Command A Reasoning", "fits": {"apple_m2_max": 5.0, "apple_m2_ultra": 10.1, "apple_m3_max": 5.0, "apple_m3_ultra": 10.3, "apple_m4_max": 6.9}, "model_id": "cohere/command-a-reasoning-08-2025", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Cohere", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8000, "cost_input": 2.5, "display_name": "Command A Translate", "fits": {"apple_m2_max": 5.0, "apple_m2_ultra": 10.1, "apple_m3_max": 5.0, "apple_m3_ultra": 10.3, "apple_m4_max": 6.9}, "model_id": "cohere/command-a-translate-08-2025", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": null}, {"benchmark_scores": {"chartqa": 70.5, "docvqa": 81.2, "mathvista": 46.5, "medqa": 73.3, "mmmu": 52.1}, "capability_tiers": {}, "context_window": 128000, "cost_input": 2.5, "display_name": "Command A Vision", "fits": {"apple_m2_max": 5.0, "apple_m2_ultra": 10.1, "apple_m3_max": 5.0, "apple_m3_ultra": 10.3, "apple_m4_max": 6.9}, "model_id": "cohere/command-a-vision-07-2025", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Cohere", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1190.0, "arena_elo_hard_prompts": 1254.6, "arena_elo_math": 1180.0, "arena_elo_overall": 1200.0, "arena_elo_style_control": 1249.1, "artificial_analysis_quality_index": 64.0, "artificial_analysis_speed_index": 78.0, "bbh": 38.5, "gpqa_diamond": 35.2, "humaneval": 65.0, "ifeval": 70.0, "math_500": 55.8, "mmlu_pro": 50.2, "musr": 12.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.15, "display_name": "Command R", "fits": {"amd_rx_7900_xtx": 38.4, "apple_m1_max": 16.0, "apple_m2_max": 16.0, "apple_m2_ultra": 32.0, "apple_m3_max": 16.0, "apple_m3_ultra": 32.8, "apple_m4": 4.8, "apple_m4_max": 21.8, "apple_m4_pro": 10.9, "nvidia_rtx_3090": 37.4, "nvidia_rtx_4090": 40.3, "nvidia_rtx_5090": 71.7}, "model_id": "cohere/command-r-08-2024", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 35.2, "arena_elo_coding": 1220.0, "arena_elo_hard_prompts": 1259.3, "arena_elo_math": 1210.0, "arena_elo_overall": 1230.0, "arena_elo_style_control": 1275.5, "artificial_analysis_quality_index": 73.0, "artificial_analysis_speed_index": 55.0, "bbh": 47.3, "bbq": 79.2, "gpqa_diamond": 42.5, "gsm8k": 70.7, "helm_safety": 83.5, "humaneval": 72.5, "ifeval": 75.0, "math_500": 65.2, "mmlu_pro": 58.5, "mt_bench": 8.3, "musr": 17.3, "toxigen": 87.5, "wildbench": 60.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 2.5, "display_name": "Command R+", "fits": {"apple_m2_max": 5.4, "apple_m2_ultra": 10.8, "apple_m3_max": 5.4, "apple_m3_ultra": 11.0, "apple_m4_max": 7.3}, "model_id": "cohere/command-r-plus-08-2024", "model_subtypes": ["llm-code"], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0375, "display_name": "Command R7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "cohere/command-r7b-12-2024", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0375, "display_name": "Command R7B Arabic", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "cohere/command-r7b-arabic-02-2025", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Cohere", "scores_as_of": null}, {"benchmark_scores": {"aider_polyglot": 58.5, "alpaca_eval": 42.8, "arena_elo_coding": 1320.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 59.1, "gsm8k": 89.3, "helm_safety": 78.2, "humaneval": 85.8, "ifeval": 84.0, "live_code_bench": 59.3, "math_500": 89.2, "mmlu_pro": 75.5, "mt_bench": 8.8, "multipl_e_csharp": 76.2, "multipl_e_julia": 56.8, "multipl_e_kotlin": 67.5, "multipl_e_lua": 50.5, "multipl_e_perl": 45.2, "multipl_e_php": 73.5, "multipl_e_r": 52.5, "multipl_e_ruby": 60.5, "multipl_e_scala": 55.8, "multipl_e_swift": 60.2, "musr": 29.8, "swe_bench_agent": 35.8, "swe_bench_verified": 42.5, "tau_bench": 41.2, "terminal_bench": 37.7, "terminal_bench_2": 46.4, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.28, "display_name": "DeepSeek Chat", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-chat", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 28.6, "gsm8k": 1.1, "hellaswag": 39.9, "mmlu_abstract_algebra": 28.0, "mmlu_anatomy": 29.6, "mmlu_astronomy": 19.7, "mmlu_business_ethics": 29.0, "mmlu_clinical_knowledge": 30.2, "mmlu_college_biology": 25.7, "mmlu_college_chemistry": 25.0, "mmlu_college_computer_science": 29.0, "mmlu_college_mathematics": 23.0, "mmlu_college_medicine": 24.3, "mmlu_college_physics": 25.5, "mmlu_computer_security": 37.0, "mmlu_conceptual_physics": 31.1, "mmlu_econometrics": 24.6, "mmlu_electrical_engineering": 38.6, "mmlu_elementary_mathematics": 25.9, "mmlu_formal_logic": 21.4, "mmlu_global_facts": 27.0, "mmlu_high_school_biology": 28.1, "mmlu_high_school_chemistry": 27.6, "mmlu_high_school_computer_science": 33.0, "mmlu_high_school_european_history": 29.1, "mmlu_high_school_geography": 35.9, "mmlu_high_school_government_and_politics": 32.6, "mmlu_high_school_macroeconomics": 26.9, "mmlu_high_school_mathematics": 24.4, "mmlu_high_school_microeconomics": 26.1, "mmlu_high_school_physics": 30.5, "mmlu_high_school_psychology": 31.7, "mmlu_high_school_statistics": 25.5, "mmlu_high_school_us_history": 25.5, "mmlu_high_school_world_history": 30.8, "mmlu_human_aging": 14.8, "mmlu_human_sexuality": 35.1, "mmlu_international_law": 28.9, "mmlu_jurisprudence": 25.0, "mmlu_logical_fallacies": 28.8, "mmlu_machine_learning": 32.1, "mmlu_management": 37.9, "mmlu_marketing": 33.3, "mmlu_medical_genetics": 21.0, "mmlu_miscellaneous": 29.2, "mmlu_moral_disputes": 28.0, "mmlu_moral_scenarios": 27.3, "mmlu_nutrition": 30.4, "mmlu_philosophy": 29.3, "mmlu_prehistory": 28.7, "mmlu_professional_accounting": 24.5, "mmlu_professional_law": 29.0, "mmlu_professional_medicine": 41.5, "mmlu_professional_psychology": 24.2, "mmlu_public_relations": 33.6, "mmlu_security_studies": 25.3, "mmlu_sociology": 33.8, "mmlu_us_foreign_policy": 32.0, "mmlu_virology": 25.3, "mmlu_world_religions": 21.6, "truthfulqa": 44.0, "winogrande": 52.4}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": 0.14, "display_name": "deepseek coder 1.3B base", "fits": {"amd_rx_7900_xt": 861.5, "amd_rx_7900_xtx": 1033.8, "amd_rx_9070_xt": 694.6, "apple_m1_max": 430.8, "apple_m2_max": 430.8, "apple_m2_ultra": 861.5, "apple_m3_max": 430.8, "apple_m3_ultra": 882.0, "apple_m4": 129.2, "apple_m4_max": 588.0, "apple_m4_pro": 294.0, "nvidia_rtx_3060_12gb": 387.7, "nvidia_rtx_3090": 1008.0, "nvidia_rtx_4060_ti_16gb": 310.2, "nvidia_rtx_4070_ti_super": 723.7, "nvidia_rtx_4080_super": 792.6, "nvidia_rtx_4090": 1085.5, "nvidia_rtx_5080": 1033.8, "nvidia_rtx_5090": 1929.8}, "model_id": "deepseek/deepseek-coder-1-3b-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 28.6, "gsm8k": 1.1, "hellaswag": 39.9, "mmlu_abstract_algebra": 28.0, "mmlu_anatomy": 29.6, "mmlu_astronomy": 19.7, "mmlu_business_ethics": 29.0, "mmlu_clinical_knowledge": 30.2, "mmlu_college_biology": 25.7, "mmlu_college_chemistry": 25.0, "mmlu_college_computer_science": 29.0, "mmlu_college_mathematics": 23.0, "mmlu_college_medicine": 24.3, "mmlu_college_physics": 25.5, "mmlu_computer_security": 37.0, "mmlu_conceptual_physics": 31.1, "mmlu_econometrics": 24.6, "mmlu_electrical_engineering": 38.6, "mmlu_elementary_mathematics": 25.9, "mmlu_formal_logic": 21.4, "mmlu_global_facts": 27.0, "mmlu_high_school_biology": 28.1, "mmlu_high_school_chemistry": 27.6, "mmlu_high_school_computer_science": 33.0, "mmlu_high_school_european_history": 29.1, "mmlu_high_school_geography": 35.9, "mmlu_high_school_government_and_politics": 32.6, "mmlu_high_school_macroeconomics": 26.9, "mmlu_high_school_mathematics": 24.4, "mmlu_high_school_microeconomics": 26.1, "mmlu_high_school_physics": 30.5, "mmlu_high_school_psychology": 31.7, "mmlu_high_school_statistics": 25.5, "mmlu_high_school_us_history": 25.5, "mmlu_high_school_world_history": 30.8, "mmlu_human_aging": 14.8, "mmlu_human_sexuality": 35.1, "mmlu_international_law": 28.9, "mmlu_jurisprudence": 25.0, "mmlu_logical_fallacies": 28.8, "mmlu_machine_learning": 32.1, "mmlu_management": 37.9, "mmlu_marketing": 33.3, "mmlu_medical_genetics": 21.0, "mmlu_miscellaneous": 29.2, "mmlu_moral_disputes": 28.0, "mmlu_moral_scenarios": 27.3, "mmlu_nutrition": 30.4, "mmlu_philosophy": 29.3, "mmlu_prehistory": 28.7, "mmlu_professional_accounting": 24.5, "mmlu_professional_law": 29.0, "mmlu_professional_medicine": 41.5, "mmlu_professional_psychology": 24.2, "mmlu_public_relations": 33.6, "mmlu_security_studies": 25.3, "mmlu_sociology": 33.8, "mmlu_us_foreign_policy": 32.0, "mmlu_virology": 25.3, "mmlu_world_religions": 21.6, "truthfulqa": 44.0, "winogrande": 52.4}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": 0.14, "display_name": "deepseek coder 1.3B instruct", "fits": {"amd_rx_7900_xt": 861.5, "amd_rx_7900_xtx": 1033.8, "amd_rx_9070_xt": 694.6, "apple_m1_max": 430.8, "apple_m2_max": 430.8, "apple_m2_ultra": 861.5, "apple_m3_max": 430.8, "apple_m3_ultra": 882.0, "apple_m4": 129.2, "apple_m4_max": 588.0, "apple_m4_pro": 294.0, "nvidia_rtx_3060_12gb": 387.7, "nvidia_rtx_3090": 1008.0, "nvidia_rtx_4060_ti_16gb": 310.2, "nvidia_rtx_4070_ti_super": 723.7, "nvidia_rtx_4080_super": 792.6, "nvidia_rtx_4090": 1085.5, "nvidia_rtx_5080": 1033.8, "nvidia_rtx_5090": 1929.8}, "model_id": "deepseek/deepseek-coder-1-3b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 37.0, "gsm8k": 18.0, "hellaswag": 53.5, "mmlu_abstract_algebra": 33.0, "mmlu_anatomy": 40.0, "mmlu_astronomy": 35.5, "mmlu_business_ethics": 40.0, "mmlu_clinical_knowledge": 41.1, "mmlu_college_biology": 30.6, "mmlu_college_chemistry": 39.0, "mmlu_college_computer_science": 43.0, "mmlu_college_mathematics": 32.0, "mmlu_college_medicine": 34.7, "mmlu_college_physics": 25.5, "mmlu_computer_security": 62.0, "mmlu_conceptual_physics": 35.3, "mmlu_econometrics": 29.8, "mmlu_electrical_engineering": 45.5, "mmlu_elementary_mathematics": 31.2, "mmlu_formal_logic": 29.4, "mmlu_global_facts": 28.0, "mmlu_high_school_biology": 36.5, "mmlu_high_school_chemistry": 29.6, "mmlu_high_school_computer_science": 52.0, "mmlu_high_school_european_history": 37.0, "mmlu_high_school_geography": 40.4, "mmlu_high_school_government_and_politics": 42.5, "mmlu_high_school_macroeconomics": 34.6, "mmlu_high_school_mathematics": 28.9, "mmlu_high_school_microeconomics": 36.1, "mmlu_high_school_physics": 27.8, "mmlu_high_school_psychology": 38.5, "mmlu_high_school_statistics": 37.0, "mmlu_high_school_us_history": 34.8, "mmlu_high_school_world_history": 32.1, "mmlu_human_aging": 38.1, "mmlu_human_sexuality": 46.6, "mmlu_international_law": 52.9, "mmlu_jurisprudence": 34.3, "mmlu_logical_fallacies": 42.3, "mmlu_machine_learning": 27.7, "mmlu_management": 42.7, "mmlu_marketing": 63.7, "mmlu_medical_genetics": 47.0, "mmlu_miscellaneous": 40.2, "mmlu_moral_disputes": 40.5, "mmlu_moral_scenarios": 28.9, "mmlu_nutrition": 40.2, "mmlu_philosophy": 44.1, "mmlu_prehistory": 28.1, "mmlu_professional_accounting": 34.8, "mmlu_professional_law": 28.8, "mmlu_professional_medicine": 44.9, "mmlu_professional_psychology": 31.7, "mmlu_public_relations": 50.9, "mmlu_security_studies": 42.9, "mmlu_sociology": 45.3, "mmlu_us_foreign_policy": 49.0, "mmlu_virology": 41.0, "mmlu_world_religions": 38.6, "truthfulqa": 40.3, "winogrande": 58.1}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": 0.14, "display_name": "deepseek coder 6.7B base", "fits": {"amd_rx_7900_xt": 167.2, "amd_rx_7900_xtx": 200.6, "amd_rx_9070_xt": 134.8, "apple_m1_max": 83.6, "apple_m2_max": 83.6, "apple_m2_ultra": 167.2, "apple_m3_max": 83.6, "apple_m3_ultra": 171.1, "apple_m4": 25.1, "apple_m4_max": 114.1, "apple_m4_pro": 57.0, "nvidia_rtx_3060_12gb": 75.2, "nvidia_rtx_3090": 195.6, "nvidia_rtx_4060_ti_16gb": 60.2, "nvidia_rtx_4070_ti_super": 140.4, "nvidia_rtx_4080_super": 153.8, "nvidia_rtx_4090": 210.6, "nvidia_rtx_5080": 200.6, "nvidia_rtx_5090": 374.4}, "model_id": "deepseek/deepseek-coder-6-7b-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 38.1, "gsm8k": 26.8, "hellaswag": 55.1, "mmlu_abstract_algebra": 33.0, "mmlu_anatomy": 34.8, "mmlu_astronomy": 33.6, "mmlu_business_ethics": 42.0, "mmlu_clinical_knowledge": 41.1, "mmlu_college_biology": 33.3, "mmlu_college_chemistry": 36.0, "mmlu_college_computer_science": 43.0, "mmlu_college_mathematics": 34.0, "mmlu_college_medicine": 33.5, "mmlu_college_physics": 23.5, "mmlu_computer_security": 61.0, "mmlu_conceptual_physics": 34.9, "mmlu_econometrics": 30.7, "mmlu_electrical_engineering": 44.1, "mmlu_elementary_mathematics": 34.1, "mmlu_formal_logic": 36.5, "mmlu_global_facts": 33.0, "mmlu_high_school_biology": 41.9, "mmlu_high_school_chemistry": 32.5, "mmlu_high_school_computer_science": 56.0, "mmlu_high_school_european_history": 37.6, "mmlu_high_school_geography": 42.4, "mmlu_high_school_government_and_politics": 43.0, "mmlu_high_school_macroeconomics": 35.6, "mmlu_high_school_mathematics": 27.8, "mmlu_high_school_microeconomics": 41.2, "mmlu_high_school_physics": 33.1, "mmlu_high_school_psychology": 41.7, "mmlu_high_school_statistics": 36.6, "mmlu_high_school_us_history": 39.7, "mmlu_high_school_world_history": 37.6, "mmlu_human_aging": 39.0, "mmlu_human_sexuality": 42.0, "mmlu_international_law": 45.5, "mmlu_jurisprudence": 37.0, "mmlu_logical_fallacies": 46.0, "mmlu_machine_learning": 33.0, "mmlu_management": 38.8, "mmlu_marketing": 65.4, "mmlu_medical_genetics": 41.0, "mmlu_miscellaneous": 43.7, "mmlu_moral_disputes": 41.9, "mmlu_moral_scenarios": 25.8, "mmlu_nutrition": 39.2, "mmlu_philosophy": 42.8, "mmlu_prehistory": 36.4, "mmlu_professional_accounting": 31.9, "mmlu_professional_law": 28.4, "mmlu_professional_medicine": 34.9, "mmlu_professional_psychology": 35.1, "mmlu_public_relations": 52.7, "mmlu_security_studies": 45.3, "mmlu_sociology": 42.8, "mmlu_us_foreign_policy": 54.0, "mmlu_virology": 37.3, "mmlu_world_religions": 35.1, "truthfulqa": 45.6, "winogrande": 56.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": 0.14, "display_name": "deepseek coder 6.7B instruct", "fits": {"amd_rx_7900_xt": 167.2, "amd_rx_7900_xtx": 200.6, "amd_rx_9070_xt": 134.8, "apple_m1_max": 83.6, "apple_m2_max": 83.6, "apple_m2_ultra": 167.2, "apple_m3_max": 83.6, "apple_m3_ultra": 171.1, "apple_m4": 25.1, "apple_m4_max": 114.1, "apple_m4_pro": 57.0, "nvidia_rtx_3060_12gb": 75.2, "nvidia_rtx_3090": 195.6, "nvidia_rtx_4060_ti_16gb": 60.2, "nvidia_rtx_4070_ti_super": 140.4, "nvidia_rtx_4080_super": 153.8, "nvidia_rtx_4090": 210.6, "nvidia_rtx_5080": 200.6, "nvidia_rtx_5090": 374.4}, "model_id": "deepseek/deepseek-coder-6-7b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.14, "display_name": "DeepSeek Coder V2 Lite Instruct", "fits": {"amd_rx_7900_xt": 70.0, "amd_rx_7900_xtx": 84.0, "amd_rx_9070_xt": 56.4, "apple_m1_max": 35.0, "apple_m2_max": 35.0, "apple_m2_ultra": 70.0, "apple_m3_max": 35.0, "apple_m3_ultra": 71.7, "apple_m4": 10.5, "apple_m4_max": 47.8, "apple_m4_pro": 23.9, "nvidia_rtx_3060_12gb": 31.5, "nvidia_rtx_3090": 81.9, "nvidia_rtx_4060_ti_16gb": 25.2, "nvidia_rtx_4070_ti_super": 58.8, "nvidia_rtx_4080_super": 64.4, "nvidia_rtx_4090": 88.2, "nvidia_rtx_5080": 84.0, "nvidia_rtx_5090": 156.8}, "model_id": "deepseek/deepseek-coder-v2-lite-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 55.7, "bbh": 35.0, "gpqa_diamond": 27.3, "gsm8k": 45.9, "hellaswag": 79.4, "ifeval": 21.8, "math_500": 2.0, "mmlu_abstract_algebra": 25.0, "mmlu_anatomy": 45.9, "mmlu_astronomy": 52.6, "mmlu_business_ethics": 55.0, "mmlu_clinical_knowledge": 53.2, "mmlu_college_biology": 50.0, "mmlu_college_chemistry": 37.0, "mmlu_college_computer_science": 46.0, "mmlu_college_mathematics": 31.0, "mmlu_college_medicine": 49.1, "mmlu_college_physics": 29.4, "mmlu_computer_security": 62.0, "mmlu_conceptual_physics": 43.4, "mmlu_econometrics": 27.2, "mmlu_electrical_engineering": 45.5, "mmlu_elementary_mathematics": 31.2, "mmlu_formal_logic": 33.3, "mmlu_global_facts": 25.0, "mmlu_high_school_biology": 57.1, "mmlu_high_school_chemistry": 35.0, "mmlu_high_school_computer_science": 47.0, "mmlu_high_school_european_history": 67.9, "mmlu_high_school_geography": 67.2, "mmlu_high_school_government_and_politics": 71.5, "mmlu_high_school_macroeconomics": 46.7, "mmlu_high_school_mathematics": 30.0, "mmlu_high_school_microeconomics": 47.5, "mmlu_high_school_physics": 29.1, "mmlu_high_school_psychology": 71.2, "mmlu_high_school_statistics": 43.5, "mmlu_high_school_us_history": 71.1, "mmlu_high_school_world_history": 72.6, "mmlu_human_aging": 57.4, "mmlu_human_sexuality": 55.7, "mmlu_international_law": 66.1, "mmlu_jurisprudence": 63.0, "mmlu_logical_fallacies": 60.7, "mmlu_machine_learning": 41.1, "mmlu_management": 65.0, "mmlu_marketing": 82.1, "mmlu_medical_genetics": 60.0, "mmlu_miscellaneous": 73.1, "mmlu_moral_disputes": 59.0, "mmlu_moral_scenarios": 30.6, "mmlu_nutrition": 52.3, "mmlu_philosophy": 55.6, "mmlu_prehistory": 56.5, "mmlu_pro": 18.1, "mmlu_professional_accounting": 36.5, "mmlu_professional_law": 40.7, "mmlu_professional_medicine": 46.7, "mmlu_professional_psychology": 52.0, "mmlu_public_relations": 59.1, "mmlu_security_studies": 62.9, "mmlu_sociology": 70.1, "mmlu_us_foreign_policy": 82.0, "mmlu_virology": 47.6, "mmlu_world_religions": 76.0, "musr": 37.4, "truthfulqa": 47.9, "winogrande": 74.9}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "deepseek llm 7B base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "deepseek/deepseek-llm-7b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 55.7, "bbh": 36.3, "gpqa_diamond": 26.6, "gsm8k": 45.9, "hellaswag": 79.4, "ifeval": 41.7, "math_500": 2.0, "mmlu_abstract_algebra": 25.0, "mmlu_anatomy": 45.9, "mmlu_astronomy": 52.6, "mmlu_business_ethics": 55.0, "mmlu_clinical_knowledge": 53.2, "mmlu_college_biology": 50.0, "mmlu_college_chemistry": 37.0, "mmlu_college_computer_science": 46.0, "mmlu_college_mathematics": 31.0, "mmlu_college_medicine": 49.1, "mmlu_college_physics": 29.4, "mmlu_computer_security": 62.0, "mmlu_conceptual_physics": 43.4, "mmlu_econometrics": 27.2, "mmlu_electrical_engineering": 45.5, "mmlu_elementary_mathematics": 31.2, "mmlu_formal_logic": 33.3, "mmlu_global_facts": 25.0, "mmlu_high_school_biology": 57.1, "mmlu_high_school_chemistry": 35.0, "mmlu_high_school_computer_science": 47.0, "mmlu_high_school_european_history": 67.9, "mmlu_high_school_geography": 67.2, "mmlu_high_school_government_and_politics": 71.5, "mmlu_high_school_macroeconomics": 46.7, "mmlu_high_school_mathematics": 30.0, "mmlu_high_school_microeconomics": 47.5, "mmlu_high_school_physics": 29.1, "mmlu_high_school_psychology": 71.2, "mmlu_high_school_statistics": 43.5, "mmlu_high_school_us_history": 71.1, "mmlu_high_school_world_history": 72.6, "mmlu_human_aging": 57.4, "mmlu_human_sexuality": 55.7, "mmlu_international_law": 66.1, "mmlu_jurisprudence": 63.0, "mmlu_logical_fallacies": 60.7, "mmlu_machine_learning": 41.1, "mmlu_management": 65.0, "mmlu_marketing": 82.1, "mmlu_medical_genetics": 60.0, "mmlu_miscellaneous": 73.1, "mmlu_moral_disputes": 59.0, "mmlu_moral_scenarios": 30.6, "mmlu_nutrition": 52.3, "mmlu_philosophy": 55.6, "mmlu_prehistory": 56.5, "mmlu_pro": 21.3, "mmlu_professional_accounting": 36.5, "mmlu_professional_law": 40.7, "mmlu_professional_medicine": 46.7, "mmlu_professional_psychology": 52.0, "mmlu_public_relations": 59.1, "mmlu_security_studies": 62.9, "mmlu_sociology": 70.1, "mmlu_us_foreign_policy": 82.0, "mmlu_virology": 47.6, "mmlu_world_religions": 76.0, "musr": 46.7, "truthfulqa": 47.9, "winogrande": 74.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "deepseek llm 7B chat", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "deepseek/deepseek-llm-7b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 34.1, "gpqa_diamond": 25.4, "ifeval": 24.5, "math_500": 2.4, "mmlu_pro": 15.1, "musr": 36.6}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "deepseek moe 16B base", "fits": {"amd_rx_7900_xt": 70.0, "amd_rx_7900_xtx": 84.0, "amd_rx_9070_xt": 56.4, "apple_m1_max": 35.0, "apple_m2_max": 35.0, "apple_m2_ultra": 70.0, "apple_m3_max": 35.0, "apple_m3_ultra": 71.7, "apple_m4": 10.5, "apple_m4_max": 47.8, "apple_m4_pro": 23.9, "nvidia_rtx_3060_12gb": 31.5, "nvidia_rtx_3090": 81.9, "nvidia_rtx_4060_ti_16gb": 25.2, "nvidia_rtx_4070_ti_super": 58.8, "nvidia_rtx_4080_super": 64.4, "nvidia_rtx_4090": 88.2, "nvidia_rtx_5080": 84.0, "nvidia_rtx_5090": 156.8}, "model_id": "deepseek/deepseek-moe-16b-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "DeepSeek OCR 2", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-ocr-2", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "DeepSeek", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "DeepSeek OCR", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-ocr", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "DeepSeek", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "DeepSeek Prover V2 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "deepseek/deepseek-prover-v2-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "DeepSeek", "scores_as_of": null}, {"benchmark_scores": {"mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": 0.55, "display_name": "DeepSeek R1 0528 Qwen3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "deepseek/deepseek-r1-0528-qwen3-8b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 65.8, "aime_2025": 82.5, "alpaca_eval": 45.2, "arena_elo_coding": 1380.0, "arena_elo_hard_prompts": 1433.6, "arena_elo_math": 1410.0, "arena_elo_overall": 1370.0, "arena_elo_style_control": 1421.7, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 68.7, "gpqa_diamond": 73.2, "humaneval": 90.5, "ifeval": 88.0, "live_code_bench": 77.0, "math_500": 97.5, "medqa": 92.1, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 81.0, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "mt_bench": 9.0, "multipl_e": 84.6, "multipl_e_cpp": 83.5, "multipl_e_csharp": 82.1, "multipl_e_go": 80.5, "multipl_e_java": 87.2, "multipl_e_javascript": 86.8, "multipl_e_julia": 64.5, "multipl_e_kotlin": 74.8, "multipl_e_lua": 57.8, "multipl_e_perl": 52.5, "multipl_e_php": 79.5, "multipl_e_python": 92.8, "multipl_e_r": 60.2, "multipl_e_ruby": 67.2, "multipl_e_rust": 75.1, "multipl_e_scala": 62.5, "multipl_e_swift": 67.5, "multipl_e_typescript": 86.1, "musr": 33.5, "swe_bench_agent": 42.5, "swe_bench_verified": 53.8, "tau_bench": 48.8, "terminal_bench": 5.7, "wildbench": 72.5}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.55, "display_name": "DeepSeek R1 0528", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-r1-0528", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1360.0, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 57.2, "gpqa_diamond": 55.2, "humaneval": 80.5, "ifeval": 78.0, "live_code_bench": 57.5, "math_500": 88.5, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 68.3, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "musr": 25.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1"}, "context_window": 131072, "cost_input": 0.88, "display_name": "DeepSeek R1 Distill Llama 70B", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "deepseek/deepseek-r1-distill-llama-70b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1360.0, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 39.8, "gpqa_diamond": 36.2, "humaneval": 62.5, "ifeval": 64.0, "math_500": 72.1, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 48.5, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "musr": 14.1}, "capability_tiers": {"reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "DeepSeek R1 Distill Llama 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "deepseek/deepseek-r1-distill-llama-8b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 32.4, "gpqa_diamond": 25.6, "ifeval": 34.6, "math_500": 16.9, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 11.9, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "musr": 36.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": 0.04, "display_name": "DeepSeek R1 Distill Qwen 1.5B", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "deepseek/deepseek-r1-distill-qwen-1-5b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1360.0, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 43.5, "gpqa_diamond": 45.3, "humaneval": 72.5, "ifeval": 72.0, "live_code_bench": 53.1, "math_500": 82.1, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 58.8, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "musr": 16.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "DeepSeek R1 Distill Qwen 14B", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "deepseek/deepseek-r1-distill-qwen-14b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1360.0, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 50.1, "gpqa_diamond": 52.8, "humaneval": 78.2, "ifeval": 76.0, "live_code_bench": 57.2, "math_500": 86.2, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 65.5, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "musr": 20.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1"}, "context_window": 131072, "cost_input": 0.3, "display_name": "DeepSeek R1 Distill Qwen 32B", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "deepseek/deepseek-r1-distill-qwen-32b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1360.0, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 38.2, "gpqa_diamond": 38.5, "humaneval": 65.0, "ifeval": 66.0, "math_500": 75.3, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 50.2, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "musr": 13.2}, "capability_tiers": {"reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "DeepSeek R1 Distill Qwen 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "deepseek/deepseek-r1-distill-qwen-7b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 65.8, "aime_2025": 79.8, "alpaca_eval": 45.2, "arena_elo_coding": 1370.0, "arena_elo_hard_prompts": 1418.1, "arena_elo_math": 1400.0, "arena_elo_overall": 1360.0, "arena_elo_style_control": 1397.5, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 68.7, "gpqa_diamond": 71.5, "humaneval": 89.2, "ifeval": 86.5, "live_code_bench": 52.1, "math_500": 97.3, "medqa": 92.1, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 79.8, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "mt_bench": 9.0, "multipl_e": 84.6, "multipl_e_cpp": 83.5, "multipl_e_csharp": 80.5, "multipl_e_go": 80.5, "multipl_e_java": 87.2, "multipl_e_javascript": 86.8, "multipl_e_julia": 62.1, "multipl_e_kotlin": 72.5, "multipl_e_lua": 55.8, "multipl_e_perl": 50.2, "multipl_e_php": 77.8, "multipl_e_python": 92.8, "multipl_e_r": 58.5, "multipl_e_ruby": 65.5, "multipl_e_rust": 75.1, "multipl_e_scala": 60.2, "multipl_e_swift": 65.2, "multipl_e_typescript": 86.1, "musr": 33.5, "swe_bench_agent": 42.5, "swe_bench_verified": 49.2, "tau_bench": 48.8, "terminal_bench": 5.7, "wildbench": 72.5}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.55, "display_name": "DeepSeek R1", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-r1", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 65.8, "aime_2025": 79.8, "alpaca_eval": 45.2, "arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1360.0, "bbh": 68.7, "gpqa_diamond": 71.5, "gsm8k": 89.3, "humaneval": 89.2, "ifeval": 86.5, "live_code_bench": 52.1, "math_500": 97.3, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 79.8, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "mt_bench": 9.0, "multipl_e": 84.6, "multipl_e_cpp": 83.5, "multipl_e_csharp": 80.5, "multipl_e_go": 80.5, "multipl_e_java": 87.2, "multipl_e_javascript": 86.8, "multipl_e_julia": 62.1, "multipl_e_kotlin": 72.5, "multipl_e_lua": 55.8, "multipl_e_perl": 50.2, "multipl_e_php": 77.8, "multipl_e_python": 92.8, "multipl_e_r": 58.5, "multipl_e_ruby": 65.5, "multipl_e_rust": 75.1, "multipl_e_scala": 60.2, "multipl_e_swift": 65.2, "multipl_e_typescript": 86.1, "musr": 33.5, "swe_bench_agent": 42.5, "swe_bench_verified": 49.2, "tau_bench": 48.8, "wildbench": 72.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.28, "display_name": "DeepSeek Reasoner", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-reasoner", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 42.8, "arena_elo_coding": 1320.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 59.1, "helm_safety": 78.2, "humaneval": 85.8, "ifeval": 84.0, "math_500": 89.2, "mmlu_pro": 75.5, "mt_bench": 8.8, "musr": 29.8, "swe_bench_verified": 42.0, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": null, "display_name": "DeepSeek V2 Lite Chat", "fits": {"amd_rx_7900_xt": 70.0, "amd_rx_7900_xtx": 84.0, "amd_rx_9070_xt": 56.4, "apple_m1_max": 35.0, "apple_m2_max": 35.0, "apple_m2_ultra": 70.0, "apple_m3_max": 35.0, "apple_m3_ultra": 71.7, "apple_m4": 10.5, "apple_m4_max": 47.8, "apple_m4_pro": 23.9, "nvidia_rtx_3060_12gb": 31.5, "nvidia_rtx_3090": 81.9, "nvidia_rtx_4060_ti_16gb": 25.2, "nvidia_rtx_4070_ti_super": 58.8, "nvidia_rtx_4080_super": 64.4, "nvidia_rtx_4090": 88.2, "nvidia_rtx_5080": 84.0, "nvidia_rtx_5090": 156.8}, "model_id": "deepseek/deepseek-v2-lite-chat", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 42.8, "arena_elo_coding": 1320.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 59.1, "helm_safety": 78.2, "humaneval": 85.8, "ifeval": 84.0, "math_500": 89.2, "mmlu_pro": 75.5, "mt_bench": 8.8, "musr": 29.8, "swe_bench_verified": 42.0, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": null, "display_name": "DeepSeek V2 Lite", "fits": {"amd_rx_7900_xt": 70.0, "amd_rx_7900_xtx": 84.0, "amd_rx_9070_xt": 56.4, "apple_m1_max": 35.0, "apple_m2_max": 35.0, "apple_m2_ultra": 70.0, "apple_m3_max": 35.0, "apple_m3_ultra": 71.7, "apple_m4": 10.5, "apple_m4_max": 47.8, "apple_m4_pro": 23.9, "nvidia_rtx_3060_12gb": 31.5, "nvidia_rtx_3090": 81.9, "nvidia_rtx_4060_ti_16gb": 25.2, "nvidia_rtx_4070_ti_super": 58.8, "nvidia_rtx_4080_super": 64.4, "nvidia_rtx_4090": 88.2, "nvidia_rtx_5080": 84.0, "nvidia_rtx_5090": 156.8}, "model_id": "deepseek/deepseek-v2-lite", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 42.8, "arena_elo_coding": 1320.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 59.1, "helm_safety": 78.2, "humaneval": 85.8, "ifeval": 84.0, "math_500": 89.2, "mmlu_pro": 75.5, "mt_bench": 8.8, "musr": 29.8, "swe_bench_verified": 42.0, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": null, "display_name": "DeepSeek V2", "fits": {"apple_m2_ultra": 4.7, "apple_m3_ultra": 4.9}, "model_id": "deepseek/deepseek-v2", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 58.5, "alpaca_eval": 42.8, "arena_elo_coding": 1330.0, "arena_elo_hard_prompts": 1408.3, "arena_elo_math": 1310.0, "arena_elo_overall": 1320.0, "arena_elo_style_control": 1394.6, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 70.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 59.1, "gsm8k": 89.3, "helm_safety": 78.2, "humaneval": 85.8, "ifeval": 84.0, "live_code_bench": 49.2, "math_500": 89.2, "mmlu_pro": 75.5, "mt_bench": 8.8, "multipl_e_csharp": 77.5, "multipl_e_julia": 58.5, "multipl_e_kotlin": 69.2, "multipl_e_lua": 52.2, "multipl_e_perl": 47.5, "multipl_e_php": 74.8, "multipl_e_r": 54.2, "multipl_e_ruby": 62.1, "multipl_e_scala": 57.2, "multipl_e_swift": 62.5, "musr": 29.8, "swe_bench_agent": 35.8, "swe_bench_verified": 42.0, "tau_bench": 41.2, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.14, "display_name": "DeepSeek V3 0324", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-v3-0324", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 58.5, "alpaca_eval": 42.8, "arena_elo_coding": 1340.0, "arena_elo_hard_prompts": 1433.3, "arena_elo_math": 1320.0, "arena_elo_overall": 1330.0, "arena_elo_style_control": 1417.9, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 70.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 62.3, "gsm8k": 89.3, "helm_safety": 78.2, "humaneval": 87.5, "ifeval": 85.5, "live_code_bench": 57.7, "math_500": 91.0, "mmlu_pro": 77.2, "mt_bench": 8.8, "multipl_e_csharp": 79.8, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.5, "multipl_e_lua": 54.5, "multipl_e_perl": 49.5, "multipl_e_php": 77.2, "multipl_e_r": 56.5, "multipl_e_ruby": 64.5, "multipl_e_scala": 59.8, "multipl_e_swift": 64.8, "musr": 29.8, "swe_bench_agent": 35.8, "swe_bench_verified": 42.5, "tau_bench": 41.2, "terminal_bench": 31.3, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.14, "display_name": "DeepSeek V3.1", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-v3-1", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 58.5, "alpaca_eval": 42.8, "arena_elo_coding": 1350.0, "arena_elo_hard_prompts": 1447.7, "arena_elo_math": 1330.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1422.8, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 70.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 64.5, "gsm8k": 89.3, "helm_safety": 78.2, "humaneval": 88.5, "ifeval": 87.0, "live_code_bench": 55.4, "math_500": 92.5, "mmlu_pro": 78.8, "mt_bench": 8.8, "multipl_e_csharp": 83.2, "multipl_e_julia": 65.5, "multipl_e_kotlin": 76.5, "multipl_e_lua": 59.2, "multipl_e_perl": 53.8, "multipl_e_php": 80.8, "multipl_e_r": 61.5, "multipl_e_ruby": 69.5, "multipl_e_scala": 64.5, "multipl_e_swift": 69.8, "musr": 29.8, "swe_bench_agent": 35.8, "swe_bench_verified": 42.5, "tau_bench": 41.2, "terminal_bench": 37.7, "terminal_bench_2": 46.4, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.14, "display_name": "DeepSeek V3.2 Exp", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-v3-2-exp", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 58.5, "alpaca_eval": 42.8, "arena_elo_coding": 1350.0, "arena_elo_hard_prompts": 1446.6, "arena_elo_math": 1330.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1424.4, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 70.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 64.5, "gsm8k": 89.3, "helm_safety": 78.2, "humaneval": 88.5, "ifeval": 87.0, "live_code_bench": 59.3, "math_500": 92.5, "mmlu_pro": 78.8, "mt_bench": 8.8, "multipl_e_csharp": 82.5, "multipl_e_julia": 64.2, "multipl_e_kotlin": 75.2, "multipl_e_lua": 58.2, "multipl_e_perl": 52.8, "multipl_e_php": 80.2, "multipl_e_r": 60.2, "multipl_e_ruby": 68.5, "multipl_e_scala": 63.2, "multipl_e_swift": 68.5, "musr": 29.8, "swe_bench_agent": 35.8, "swe_bench_verified": 42.5, "tau_bench": 41.2, "terminal_bench": 37.7, "terminal_bench_2": 46.4, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.14, "display_name": "DeepSeek V3.2", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-v3-2", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 58.5, "alpaca_eval": 42.8, "arena_elo_coding": 1330.0, "arena_elo_hard_prompts": 1350.3, "arena_elo_math": 1310.0, "arena_elo_overall": 1320.0, "arena_elo_style_control": 1358.2, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 70.0, "bbh": 64.1, "bbq": 73.5, "gpqa_diamond": 59.1, "gsm8k": 89.3, "helm_safety": 78.2, "humaneval": 85.8, "ifeval": 84.0, "live_code_bench": 40.5, "math_500": 89.2, "medqa": 80.3, "mmlu_pro": 75.5, "mt_bench": 8.8, "multipl_e_csharp": 76.2, "multipl_e_julia": 56.8, "multipl_e_kotlin": 67.5, "multipl_e_lua": 50.5, "multipl_e_perl": 45.2, "multipl_e_php": 73.5, "multipl_e_r": 52.5, "multipl_e_ruby": 60.5, "multipl_e_scala": 55.8, "multipl_e_swift": 60.2, "musr": 29.8, "swe_bench_agent": 35.8, "swe_bench_verified": 42.0, "tau_bench": 41.2, "terminal_bench": 37.7, "terminal_bench_2": 46.4, "toxigen": 82.5, "wildbench": 70.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.14, "display_name": "DeepSeek V3", "fits": {"apple_m3_ultra": 1.7}, "model_id": "deepseek/deepseek-v3", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "DeepSeek", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "deepseek vl2 tiny", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "deepseek/deepseek-vl2-tiny", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "DeepSeek", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Janus Pro 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "deepseek/janus-pro-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "DeepSeek", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "bert for seq generation L 24 bbc encoder", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "google/bert-for-seq-generation-l-24-bbc-encoder", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "bert uncased L 2 H 128 A 2", "fits": {"amd_rx_7900_xt": 254545.5, "amd_rx_7900_xtx": 305454.5, "amd_rx_9070_xt": 205227.3, "apple_m1_max": 127272.7, "apple_m2_max": 127272.7, "apple_m2_ultra": 254545.5, "apple_m3_max": 127272.7, "apple_m3_ultra": 260590.9, "apple_m4": 38181.8, "apple_m4_max": 173727.3, "apple_m4_pro": 86863.6, "nvidia_rtx_3060_12gb": 114545.5, "nvidia_rtx_3090": 297818.2, "nvidia_rtx_4060_ti_16gb": 91636.4, "nvidia_rtx_4070_ti_super": 213818.2, "nvidia_rtx_4080_super": 234181.8, "nvidia_rtx_4090": 320727.3, "nvidia_rtx_5080": 305454.5, "nvidia_rtx_5090": 570181.8}, "model_id": "google/bert-uncased-l-2-h-128-a-2", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "bert uncased L 4 H 256 A 4", "fits": {"amd_rx_7900_xt": 99115.0, "amd_rx_7900_xtx": 118938.1, "amd_rx_9070_xt": 79911.5, "apple_m1_max": 49557.5, "apple_m2_max": 49557.5, "apple_m2_ultra": 99115.0, "apple_m3_max": 49557.5, "apple_m3_ultra": 101469.0, "apple_m4": 14867.3, "apple_m4_max": 67646.0, "apple_m4_pro": 33823.0, "nvidia_rtx_3060_12gb": 44601.8, "nvidia_rtx_3090": 115964.6, "nvidia_rtx_4060_ti_16gb": 35681.4, "nvidia_rtx_4070_ti_super": 83256.6, "nvidia_rtx_4080_super": 91185.8, "nvidia_rtx_4090": 124885.0, "nvidia_rtx_5080": 118938.1, "nvidia_rtx_5090": 222017.7}, "model_id": "google/bert-uncased-l-4-h-256-a-4", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "bigbird roberta base", "fits": {"amd_rx_7900_xt": 8750.0, "amd_rx_7900_xtx": 10500.0, "amd_rx_9070_xt": 7054.7, "apple_m1_max": 4375.0, "apple_m2_max": 4375.0, "apple_m2_ultra": 8750.0, "apple_m3_max": 4375.0, "apple_m3_ultra": 8957.8, "apple_m4": 1312.5, "apple_m4_max": 5971.9, "apple_m4_pro": 2985.9, "nvidia_rtx_3060_12gb": 3937.5, "nvidia_rtx_3090": 10237.5, "nvidia_rtx_4060_ti_16gb": 3150.0, "nvidia_rtx_4070_ti_super": 7350.0, "nvidia_rtx_4080_super": 8050.0, "nvidia_rtx_4090": 11025.0, "nvidia_rtx_5080": 10500.0, "nvidia_rtx_5090": 19600.0}, "model_id": "google/bigbird-roberta-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "byt5 base", "fits": {"amd_rx_7900_xt": 1931.0, "amd_rx_7900_xtx": 2317.2, "amd_rx_9070_xt": 1556.9, "apple_m1_max": 965.5, "apple_m2_max": 965.5, "apple_m2_ultra": 1931.0, "apple_m3_max": 965.5, "apple_m3_ultra": 1976.9, "apple_m4": 289.7, "apple_m4_max": 1317.9, "apple_m4_pro": 659.0, "nvidia_rtx_3060_12gb": 869.0, "nvidia_rtx_3090": 2259.3, "nvidia_rtx_4060_ti_16gb": 695.2, "nvidia_rtx_4070_ti_super": 1622.1, "nvidia_rtx_4080_super": 1776.6, "nvidia_rtx_4090": 2433.1, "nvidia_rtx_5080": 2317.2, "nvidia_rtx_5090": 4325.5}, "model_id": "google/byt5-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "byt5 small", "fits": {"amd_rx_7900_xt": 3733.3, "amd_rx_7900_xtx": 4480.0, "amd_rx_9070_xt": 3010.0, "apple_m1_max": 1866.7, "apple_m2_max": 1866.7, "apple_m2_ultra": 3733.3, "apple_m3_max": 1866.7, "apple_m3_ultra": 3822.0, "apple_m4": 560.0, "apple_m4_max": 2548.0, "apple_m4_pro": 1274.0, "nvidia_rtx_3060_12gb": 1680.0, "nvidia_rtx_3090": 4368.0, "nvidia_rtx_4060_ti_16gb": 1344.0, "nvidia_rtx_4070_ti_super": 3136.0, "nvidia_rtx_4080_super": 3434.7, "nvidia_rtx_4090": 4704.0, "nvidia_rtx_5080": 4480.0, "nvidia_rtx_5090": 8362.7}, "model_id": "google/byt5-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "electra base discriminator", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "google/electra-base-discriminator", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "electra small discriminator", "fits": {"amd_rx_7900_xt": 80000.0, "amd_rx_7900_xtx": 96000.0, "amd_rx_9070_xt": 64500.0, "apple_m1_max": 40000.0, "apple_m2_max": 40000.0, "apple_m2_ultra": 80000.0, "apple_m3_max": 40000.0, "apple_m3_ultra": 81900.0, "apple_m4": 12000.0, "apple_m4_max": 54600.0, "apple_m4_pro": 27300.0, "nvidia_rtx_3060_12gb": 36000.0, "nvidia_rtx_3090": 93600.0, "nvidia_rtx_4060_ti_16gb": 28800.0, "nvidia_rtx_4070_ti_super": 67200.0, "nvidia_rtx_4080_super": 73600.0, "nvidia_rtx_4090": 100800.0, "nvidia_rtx_5080": 96000.0, "nvidia_rtx_5090": 179200.0}, "model_id": "google/electra-small-discriminator", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "embeddinggemma 300M", "fits": {"amd_rx_7900_xt": 3733.3, "amd_rx_7900_xtx": 4480.0, "amd_rx_9070_xt": 3010.0, "apple_m1_max": 1866.7, "apple_m2_max": 1866.7, "apple_m2_ultra": 3733.3, "apple_m3_max": 1866.7, "apple_m3_ultra": 3822.0, "apple_m4": 560.0, "apple_m4_max": 2548.0, "apple_m4_pro": 1274.0, "nvidia_rtx_3060_12gb": 1680.0, "nvidia_rtx_3090": 4368.0, "nvidia_rtx_4060_ti_16gb": 1344.0, "nvidia_rtx_4070_ti_super": 3136.0, "nvidia_rtx_4080_super": 3434.7, "nvidia_rtx_4090": 4704.0, "nvidia_rtx_5080": 4480.0, "nvidia_rtx_5090": 8362.7}, "model_id": "google/embeddinggemma-300m", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"bbh": 35.3, "gpqa_diamond": 23.8, "ifeval": 18.9, "math_500": 1.1, "mmlu_pro": 13.6, "musr": 36.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "flan t5 base", "fits": {"amd_rx_7900_xt": 4480.0, "amd_rx_7900_xtx": 5376.0, "amd_rx_9070_xt": 3612.0, "apple_m1_max": 2240.0, "apple_m2_max": 2240.0, "apple_m2_ultra": 4480.0, "apple_m3_max": 2240.0, "apple_m3_ultra": 4586.4, "apple_m4": 672.0, "apple_m4_max": 3057.6, "apple_m4_pro": 1528.8, "nvidia_rtx_3060_12gb": 2016.0, "nvidia_rtx_3090": 5241.6, "nvidia_rtx_4060_ti_16gb": 1612.8, "nvidia_rtx_4070_ti_super": 3763.2, "nvidia_rtx_4080_super": 4121.6, "nvidia_rtx_4090": 5644.8, "nvidia_rtx_5080": 5376.0, "nvidia_rtx_5090": 10035.2}, "model_id": "google/flan-t5-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 41.5, "gpqa_diamond": 25.1, "ifeval": 22.0, "math_500": 1.4, "mmlu_pro": 17.1, "musr": 40.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "flan t5 large", "fits": {"amd_rx_7900_xt": 1435.9, "amd_rx_7900_xtx": 1723.1, "amd_rx_9070_xt": 1157.7, "apple_m1_max": 717.9, "apple_m2_max": 717.9, "apple_m2_ultra": 1435.9, "apple_m3_max": 717.9, "apple_m3_ultra": 1470.0, "apple_m4": 215.4, "apple_m4_max": 980.0, "apple_m4_pro": 490.0, "nvidia_rtx_3060_12gb": 646.2, "nvidia_rtx_3090": 1680.0, "nvidia_rtx_4060_ti_16gb": 516.9, "nvidia_rtx_4070_ti_super": 1206.2, "nvidia_rtx_4080_super": 1321.0, "nvidia_rtx_4090": 1809.2, "nvidia_rtx_5080": 1723.1, "nvidia_rtx_5090": 3216.4}, "model_id": "google/flan-t5-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 32.8, "gpqa_diamond": 26.1, "ifeval": 15.2, "math_500": 0.8, "mmlu_pro": 12.3, "musr": 41.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "flan t5 small", "fits": {"amd_rx_7900_xt": 18666.7, "amd_rx_7900_xtx": 22400.0, "amd_rx_9070_xt": 15050.0, "apple_m1_max": 9333.3, "apple_m2_max": 9333.3, "apple_m2_ultra": 18666.7, "apple_m3_max": 9333.3, "apple_m3_ultra": 19110.0, "apple_m4": 2800.0, "apple_m4_max": 12740.0, "apple_m4_pro": 6370.0, "nvidia_rtx_3060_12gb": 8400.0, "nvidia_rtx_3090": 21840.0, "nvidia_rtx_4060_ti_16gb": 6720.0, "nvidia_rtx_4070_ti_super": 15680.0, "nvidia_rtx_4080_super": 17173.3, "nvidia_rtx_4090": 23520.0, "nvidia_rtx_5080": 22400.0, "nvidia_rtx_5090": 41813.3}, "model_id": "google/flan-t5-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 45.4, "gpqa_diamond": 24.6, "ifeval": 22.1, "math_500": 0.1, "mmlu_pro": 21.4, "musr": 42.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "flan t5 xl", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "google/flan-t5-xl", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "fnet base", "fits": {"amd_rx_7900_xt": 13658.5, "amd_rx_7900_xtx": 16390.2, "amd_rx_9070_xt": 11012.2, "apple_m1_max": 6829.3, "apple_m2_max": 6829.3, "apple_m2_ultra": 13658.5, "apple_m3_max": 6829.3, "apple_m3_ultra": 13982.9, "apple_m4": 2048.8, "apple_m4_max": 9322.0, "apple_m4_pro": 4661.0, "nvidia_rtx_3060_12gb": 6146.3, "nvidia_rtx_3090": 15980.5, "nvidia_rtx_4060_ti_16gb": 4917.1, "nvidia_rtx_4070_ti_super": 11473.2, "nvidia_rtx_4080_super": 12565.9, "nvidia_rtx_4090": 17209.8, "nvidia_rtx_5080": 16390.2, "nvidia_rtx_5090": 30595.1}, "model_id": "google/fnet-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1220.0, "arena_elo_hard_prompts": 1258.2, "arena_elo_math": 1210.0, "arena_elo_overall": 1230.0, "arena_elo_style_control": 1258.1, "arena_elo_vision": 1070.6, "artificial_analysis_quality_index": 71.0, "artificial_analysis_speed_index": 86.0, "bbq": 81.5, "chartqa": 74.8, "docvqa": 85.2, "flores_en_de": 63.2, "flores_en_es": 68.1, "flores_en_ja": 58.1, "flores_en_zh": 56.8, "gpqa_diamond": 45.2, "gsm8k": 86.2, "helm_safety": 86.2, "humaneval": 78.5, "ifeval": 77.0, "math_500": 65.8, "mathvista": 54.5, "mgsm": 82.6, "mmlu_pro": 63.5, "mmmu": 56.1, "toxigen": 90.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.0375, "display_name": "Gemini 1.5 Flash-8B", "fits": {}, "model_id": "google/gemini-1-5-flash-8b", "model_subtypes": ["llm-code"], "model_type": "vlm", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1220.0, "arena_elo_hard_prompts": 1302.4, "arena_elo_math": 1210.0, "arena_elo_overall": 1230.0, "arena_elo_style_control": 1309.1, "arena_elo_vision": 1140.1, "artificial_analysis_quality_index": 71.0, "artificial_analysis_speed_index": 86.0, "bbq": 81.5, "chartqa": 74.8, "docvqa": 85.2, "flores_en_de": 63.2, "flores_en_es": 68.1, "flores_en_ja": 58.1, "flores_en_zh": 56.8, "gpqa_diamond": 45.2, "gsm8k": 86.2, "helm_safety": 86.2, "humaneval": 78.5, "ifeval": 77.0, "math_500": 65.8, "mathvista": 54.5, "mgsm": 82.6, "mmlu_pro": 63.5, "mmmu": 56.1, "toxigen": 90.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.075, "display_name": "Gemini 1.5 Flash", "fits": {}, "model_id": "google/gemini-1-5-flash", "model_subtypes": ["llm-code"], "model_type": "vlm", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 82.5, "alpaca_eval": 24.4, "arena_elo_coding": 1250.0, "arena_elo_hard_prompts": 1350.2, "arena_elo_math": 1240.0, "arena_elo_overall": 1260.0, "arena_elo_style_control": 1350.6, "arena_elo_vision": 1178.6, "artificial_analysis_quality_index": 78.0, "artificial_analysis_speed_index": 68.0, "bbq": 82.8, "chartqa": 79.2, "docvqa": 88.5, "flores_en_de": 63.2, "flores_en_es": 68.1, "flores_en_ja": 58.1, "flores_en_zh": 56.8, "gpqa_diamond": 55.8, "gsm8k": 90.8, "helm_safety": 87.5, "humaneval": 84.1, "ifeval": 82.0, "math_500": 74.3, "mathvista": 55.8, "mgsm": 87.5, "mmlu_pro": 72.0, "mmmu": 62.8, "toxigen": 91.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 1.25, "display_name": "Gemini 1.5 Pro", "fits": {}, "model_id": "google/gemini-1-5-pro", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 78.5, "arena_elo_coding": 1260.0, "arena_elo_hard_prompts": 1347.9, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "arena_elo_style_control": 1352.8, "arena_elo_vision": 1134.2, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 93.0, "bbq": 80.8, "chartqa": 78.5, "docvqa": 88.1, "gpqa_diamond": 48.5, "helm_safety": 85.5, "humaneval": 80.2, "ifeval": 78.5, "live_code_bench": 35.1, "math_500": 72.1, "mathvista": 55.2, "mmlu_pro": 64.8, "mmmu": 62.5, "toxigen": 89.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.075, "display_name": "Gemini 2.0 Flash Lite", "fits": {}, "model_id": "google/gemini-2-0-flash-lite", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 83.3, "arena_elo_coding": 1300.0, "arena_elo_hard_prompts": 1360.4, "arena_elo_math": 1290.0, "arena_elo_overall": 1310.0, "arena_elo_style_control": 1360.0, "arena_elo_vision": 1170.6, "artificial_analysis_quality_index": 77.0, "artificial_analysis_speed_index": 89.0, "bbq": 83.2, "chartqa": 78.5, "docvqa": 88.1, "flores_en_de": 63.2, "flores_en_es": 68.1, "flores_en_ja": 58.1, "flores_en_zh": 56.8, "gpqa_diamond": 60.1, "gsm8k": 86.2, "helm_safety": 87.8, "humaneval": 85.5, "ifeval": 84.0, "live_code_bench": 35.1, "math_500": 82.3, "mathvista": 55.2, "medqa": 83.2, "mgsm": 82.6, "mmlu_pro": 73.5, "mmmu": 62.5, "ocrbench": 68.6, "toxigen": 91.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.1, "display_name": "Gemini 2.0 Flash", "fits": {}, "model_id": "google/gemini-2-0-flash", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 49.5, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1"}, "context_window": 32768, "cost_input": 0.3, "display_name": "Gemini 2.5 Flash Image (Preview)", "fits": {}, "model_id": "google/gemini-2-5-flash-image-preview", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "image-generation", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 49.5, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1"}, "context_window": 32768, "cost_input": 0.3, "display_name": "Gemini 2.5 Flash Image", "fits": {}, "model_id": "google/gemini-2-5-flash-image", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "image-generation", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 33.7, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.1, "display_name": "Gemini 2.5 Flash Lite Preview 06-17", "fits": {}, "model_id": "google/gemini-2-5-flash-lite-preview-06-17", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_hard_prompts": 1390.7, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1380.0, "arena_elo_vision": 1173.2, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 33.7, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.1, "display_name": "Gemini 2.5 Flash Lite Preview 09-25", "fits": {}, "model_id": "google/gemini-2-5-flash-lite-preview-09-2025", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_hard_prompts": 1380.9, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1374.3, "arena_elo_vision": 1188.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 33.7, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.1, "display_name": "Gemini 2.5 Flash Lite", "fits": {}, "model_id": "google/gemini-2-5-flash-lite", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 47.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 49.5, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.15, "display_name": "Gemini 2.5 Flash Preview 04-17", "fits": {}, "model_id": "google/gemini-2-5-flash-preview-04-17", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 49.5, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.15, "display_name": "Gemini 2.5 Flash Preview 05-20", "fits": {}, "model_id": "google/gemini-2-5-flash-preview-05-20", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_hard_prompts": 1420.7, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1405.1, "arena_elo_vision": 1226.1, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 49.5, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.3, "display_name": "Gemini 2.5 Flash Preview 09-25", "fits": {}, "model_id": "google/gemini-2-5-flash-preview-09-2025", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "gpqa_diamond": 70.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 49.5, "math_500": 90.5, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1"}, "context_window": 8000, "cost_input": 0.5, "display_name": "Gemini 2.5 Flash Preview TTS", "fits": {}, "model_id": "google/gemini-2-5-flash-preview-tts", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "audio-tts", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 55.1, "aime_2025": 73.3, "arena_elo_coding": 1330.0, "arena_elo_hard_prompts": 1420.0, "arena_elo_math": 1350.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1411.0, "arena_elo_vision": 1213.4, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 87.0, "flores_en_de": 63.2, "flores_en_es": 68.1, "flores_en_ja": 58.1, "flores_en_zh": 56.8, "gpqa_diamond": 70.2, "gsm8k": 86.2, "humaneval": 88.8, "ifeval": 87.5, "live_code_bench": 49.5, "math_500": 90.5, "mgsm": 82.6, "mmlu_pro": 78.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.5, "multipl_e_kotlin": 71.5, "multipl_e_lua": 53.8, "multipl_e_perl": 49.5, "multipl_e_php": 76.2, "multipl_e_r": 56.2, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.8, "swe_bench_verified": 49.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.3, "display_name": "Gemini 2.5 Flash", "fits": {}, "model_id": "google/gemini-2-5-flash", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 95.8, "aider_polyglot": 76.9, "aime_2025": 86.7, "alpaca_eval": 51.5, "arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1380.0, "artificial_analysis_quality_index": 85.0, "artificial_analysis_speed_index": 71.0, "bbq": 84.2, "chartqa": 87.5, "docvqa": 93.8, "finbench": 69.8, "gpqa_diamond": 80.5, "helm_safety": 88.5, "humaneval": 91.2, "ifeval": 91.0, "legalbench": 73.8, "live_code_bench": 57.8, "math_500": 95.2, "mathvista": 67.8, "medqa": 80.2, "mmlu_pro": 84.0, "mmmu": 72.1, "mt_bench": 9.2, "ocrbench": 85.2, "swe_bench_verified": 63.8, "toxigen": 91.5, "wildbench": 79.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 1.25, "display_name": "Gemini 2.5 Pro Preview 05-06", "fits": {}, "model_id": "google/gemini-2-5-pro-preview-05-06", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 95.8, "aider_polyglot": 83.1, "aime_2025": 86.7, "alpaca_eval": 51.5, "arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1380.0, "artificial_analysis_quality_index": 85.0, "artificial_analysis_speed_index": 71.0, "bbq": 84.2, "chartqa": 87.5, "docvqa": 93.8, "finbench": 69.8, "gpqa_diamond": 80.5, "helm_safety": 88.5, "humaneval": 91.2, "ifeval": 91.0, "legalbench": 73.8, "live_code_bench": 57.8, "math_500": 95.2, "mathvista": 67.8, "medqa": 80.2, "mmlu_pro": 84.0, "mmmu": 72.1, "mt_bench": 9.2, "ocrbench": 85.2, "swe_bench_verified": 63.8, "toxigen": 91.5, "wildbench": 79.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 1.25, "display_name": "Gemini 2.5 Pro Preview 06-05", "fits": {}, "model_id": "google/gemini-2-5-pro-preview-06-05", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 95.8, "aider_polyglot": 72.9, "aime_2025": 86.7, "alpaca_eval": 51.5, "arena_elo_coding": 1370.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1380.0, "artificial_analysis_quality_index": 85.0, "artificial_analysis_speed_index": 71.0, "bbq": 84.2, "chartqa": 87.5, "docvqa": 93.8, "finbench": 69.8, "gpqa_diamond": 80.5, "helm_safety": 88.5, "humaneval": 91.2, "ifeval": 91.0, "legalbench": 73.8, "live_code_bench": 57.8, "math_500": 95.2, "mathvista": 67.8, "medqa": 80.2, "mmlu_pro": 84.0, "mmmu": 72.1, "mt_bench": 9.2, "ocrbench": 85.2, "swe_bench_verified": 63.8, "toxigen": 91.5, "wildbench": 79.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1"}, "context_window": 8000, "cost_input": 1.0, "display_name": "Gemini 2.5 Pro Preview TTS", "fits": {}, "model_id": "google/gemini-2-5-pro-preview-tts", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "audio-tts", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 95.8, "aider_polyglot": 71.2, "aime_2025": 86.7, "alpaca_eval": 51.5, "arena_elo_coding": 1370.0, "arena_elo_hard_prompts": 1460.5, "arena_elo_math": 1400.0, "arena_elo_overall": 1380.0, "arena_elo_style_control": 1448.2, "arena_elo_vision": 1246.0, "artificial_analysis_quality_index": 85.0, "artificial_analysis_speed_index": 71.0, "bbq": 84.2, "chartqa": 87.5, "docvqa": 93.8, "finbench": 69.8, "flores_en_de": 63.2, "flores_en_es": 68.1, "flores_en_ja": 58.1, "flores_en_zh": 56.8, "gpqa_diamond": 80.5, "gsm8k": 90.8, "helm_safety": 88.5, "humaneval": 91.2, "ifeval": 91.0, "legalbench": 73.8, "live_code_bench": 57.8, "math_500": 95.2, "mathvista": 67.8, "medqa": 80.2, "mgsm": 87.5, "mmlu_astronomy": 79.8, "mmlu_biology": 87.2, "mmlu_business_ethics": 81.2, "mmlu_chemistry": 81.5, "mmlu_clinical_knowledge": 86.5, "mmlu_computer_science": 87.8, "mmlu_jurisprudence": 79.2, "mmlu_physics": 84.8, "mmlu_pro": 84.0, "mmlu_professional_accounting": 71.2, "mmlu_professional_law": 77.2, "mmmu": 72.1, "mt_bench": 9.2, "multipl_e": 84.9, "multipl_e_cpp": 84.2, "multipl_e_csharp": 84.8, "multipl_e_go": 81.2, "multipl_e_java": 87.8, "multipl_e_javascript": 87.5, "multipl_e_julia": 67.2, "multipl_e_kotlin": 79.2, "multipl_e_lua": 61.8, "multipl_e_perl": 58.5, "multipl_e_php": 82.5, "multipl_e_python": 91.8, "multipl_e_r": 64.8, "multipl_e_ruby": 74.2, "multipl_e_rust": 75.5, "multipl_e_scala": 68.5, "multipl_e_swift": 72.5, "multipl_e_typescript": 86.5, "ocrbench": 85.2, "swe_bench_agent": 55.5, "swe_bench_verified": 63.8, "tau_bench": 58.2, "terminal_bench": 45.5, "toxigen": 91.5, "wildbench": 79.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 1048576, "cost_input": 1.25, "display_name": "Gemini 2.5 Pro", "fits": {}, "model_id": "google/gemini-2-5-pro", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2"}, "context_window": 131072, "cost_input": 0.25, "display_name": "Gemini 3.1 Flash Image (Preview)", "fits": {}, "model_id": "google/gemini-3-1-flash-image-preview", "model_subtypes": [], "model_type": "image-generation", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.25, "display_name": "Gemini 3.1 Flash Lite Preview", "fits": {}, "model_id": "google/gemini-3-1-flash-lite-preview", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 2.0, "display_name": "Gemini 3.1 Pro Preview Custom Tools", "fits": {}, "model_id": "google/gemini-3-1-pro-preview-customtools", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"gpqa_diamond": 94.3, "hle": 44.4, "hle_tools": 51.4, "legalbench": 87.4, "medqa": 96.4, "swe_bench_pro": 54.2, "swe_bench_verified": 80.6, "terminal_bench_2": 68.5, "usamo_2026": 74.4}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 2.0, "display_name": "Gemini 3.1 Pro Preview", "fits": {}, "model_id": "google/gemini-3-1-pro-preview", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"legalbench": 86.9, "live_code_bench": 79.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.5, "display_name": "Gemini 3 Flash Preview", "fits": {}, "model_id": "google/gemini-3-flash-preview", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"legalbench": 87.0, "live_code_bench": 91.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 2.0, "display_name": "Gemini 3 Pro Preview", "fits": {}, "model_id": "google/gemini-3-pro-preview", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 55.5, "miracl": 56.2, "mteb_classification": 70.8, "mteb_clustering": 50.5, "mteb_overall": 66.2, "mteb_pair_classification": 85.1, "mteb_reranking": 59.5, "mteb_retrieval": 63.5, "mteb_sts": 84.5, "mteb_summarization": 30.5}, "capability_tiers": {}, "context_window": 2048, "cost_input": 0.15, "display_name": "Gemini Embedding 001", "fits": {}, "model_id": "google/gemini-embedding-001", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.3, "display_name": "Gemini Flash Latest", "fits": {}, "model_id": "google/gemini-flash-latest", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 0.1, "display_name": "Gemini Flash-Lite Latest", "fits": {}, "model_id": "google/gemini-flash-lite-latest", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.5, "display_name": "Gemini Live 2.5 Flash Preview Native Audio", "fits": {}, "model_id": "google/gemini-live-2-5-flash-preview-native-audio", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.5, "display_name": "Gemini Live 2.5 Flash", "fits": {}, "model_id": "google/gemini-live-2-5-flash", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"bbh": 31.8, "gpqa_diamond": 26.9, "ifeval": 30.7, "math_500": 1.8, "mmlu_pro": 14.8, "musr": 33.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "gemma 1.1 2B it", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/gemma-1-1-2b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {"arena_elo_coding": 1230.0, "arena_elo_hard_prompts": 1280.9, "arena_elo_math": 1210.0, "arena_elo_overall": 1220.0, "arena_elo_style_control": 1287.6, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 80.0, "bbh": 52.4, "gpqa_diamond": 48.1, "gsm8k": 74.0, "humaneval": 72.5, "ifeval": 75.0, "math_500": 68.5, "mmlu_pro": 60.2, "musr": 20.1}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.15, "display_name": "gemma 2 27B it", "fits": {"amd_rx_7900_xt": 41.5, "amd_rx_7900_xtx": 49.8, "apple_m1_max": 20.7, "apple_m2_max": 20.7, "apple_m2_ultra": 41.5, "apple_m3_max": 20.7, "apple_m3_ultra": 42.5, "apple_m4": 6.2, "apple_m4_max": 28.3, "apple_m4_pro": 14.2, "nvidia_rtx_3090": 48.5, "nvidia_rtx_4090": 52.3, "nvidia_rtx_5090": 92.9}, "model_id": "google/gemma-2-27b-it", "model_subtypes": ["llm-code"], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 44.0, "gpqa_diamond": 29.7, "ifeval": 12.7, "math_500": 3.4, "mmlu_pro": 27.2, "musr": 42.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.03, "display_name": "gemma 2 2B it", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/gemma-2-2b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 37.1, "gpqa_diamond": 26.3, "ifeval": 20.2, "math_500": 3.0, "mmlu_pro": 22.2, "musr": 42.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.03, "display_name": "gemma 2 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/gemma-2-2b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {"arena_elo_coding": 1200.0, "arena_elo_hard_prompts": 1256.0, "arena_elo_math": 1180.0, "arena_elo_overall": 1190.0, "arena_elo_style_control": 1265.0, "artificial_analysis_quality_index": 60.0, "artificial_analysis_speed_index": 90.0, "bbh": 42.8, "gpqa_diamond": 38.2, "gsm8k": 68.6, "humaneval": 60.5, "ifeval": 68.0, "math_500": 52.1, "mmlu_pro": 48.5, "musr": 15.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.08, "display_name": "gemma 2 9B it", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "google/gemma-2-9b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1200.0, "arena_elo_math": 1180.0, "arena_elo_overall": 1190.0, "artificial_analysis_quality_index": 60.0, "artificial_analysis_speed_index": 90.0, "bbh": 42.8, "gpqa_diamond": 38.2, "gsm8k": 68.6, "humaneval": 60.5, "ifeval": 68.0, "math_500": 52.1, "mmlu_pro": 48.5, "musr": 15.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.08, "display_name": "gemma 2 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "google/gemma-2-9b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 43.9, "bbh": 31.5, "gpqa_diamond": 27.9, "gsm8k": 5.5, "hellaswag": 62.7, "ifeval": 26.9, "math_500": 2.0, "mmlu_abstract_algebra": 28.0, "mmlu_anatomy": 38.5, "mmlu_astronomy": 33.6, "mmlu_business_ethics": 48.0, "mmlu_clinical_knowledge": 42.3, "mmlu_college_biology": 34.0, "mmlu_college_chemistry": 26.0, "mmlu_college_computer_science": 31.0, "mmlu_college_mathematics": 26.0, "mmlu_college_medicine": 35.8, "mmlu_college_physics": 18.6, "mmlu_computer_security": 45.0, "mmlu_conceptual_physics": 35.3, "mmlu_econometrics": 28.1, "mmlu_electrical_engineering": 46.2, "mmlu_elementary_mathematics": 24.9, "mmlu_formal_logic": 25.4, "mmlu_global_facts": 30.0, "mmlu_high_school_biology": 31.6, "mmlu_high_school_chemistry": 29.1, "mmlu_high_school_computer_science": 37.0, "mmlu_high_school_european_history": 46.1, "mmlu_high_school_geography": 46.0, "mmlu_high_school_government_and_politics": 47.7, "mmlu_high_school_macroeconomics": 32.8, "mmlu_high_school_mathematics": 20.0, "mmlu_high_school_microeconomics": 34.0, "mmlu_high_school_physics": 25.8, "mmlu_high_school_psychology": 51.0, "mmlu_high_school_statistics": 20.4, "mmlu_high_school_us_history": 42.6, "mmlu_high_school_world_history": 51.9, "mmlu_human_aging": 39.0, "mmlu_human_sexuality": 42.7, "mmlu_international_law": 52.1, "mmlu_jurisprudence": 46.3, "mmlu_logical_fallacies": 36.2, "mmlu_machine_learning": 33.0, "mmlu_management": 44.7, "mmlu_marketing": 59.4, "mmlu_medical_genetics": 39.0, "mmlu_miscellaneous": 46.2, "mmlu_moral_disputes": 40.8, "mmlu_moral_scenarios": 25.3, "mmlu_nutrition": 45.1, "mmlu_philosophy": 40.2, "mmlu_prehistory": 40.7, "mmlu_pro": 13.5, "mmlu_professional_accounting": 30.5, "mmlu_professional_law": 31.9, "mmlu_professional_medicine": 20.2, "mmlu_professional_psychology": 38.2, "mmlu_public_relations": 41.8, "mmlu_security_studies": 47.3, "mmlu_sociology": 43.8, "mmlu_us_foreign_policy": 64.0, "mmlu_virology": 41.0, "mmlu_world_religions": 43.9, "musr": 33.4, "truthfulqa": 45.8, "winogrande": 60.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "gemma 2B it", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/gemma-2b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 48.4, "bbh": 33.7, "gpqa_diamond": 25.5, "gsm8k": 16.9, "hellaswag": 71.8, "ifeval": 20.4, "math_500": 3.0, "mmlu_abstract_algebra": 26.0, "mmlu_anatomy": 48.9, "mmlu_astronomy": 42.8, "mmlu_business_ethics": 44.0, "mmlu_clinical_knowledge": 46.8, "mmlu_college_biology": 45.8, "mmlu_college_chemistry": 37.0, "mmlu_college_computer_science": 38.0, "mmlu_college_mathematics": 30.0, "mmlu_college_medicine": 42.8, "mmlu_college_physics": 14.7, "mmlu_computer_security": 53.0, "mmlu_conceptual_physics": 41.7, "mmlu_econometrics": 31.6, "mmlu_electrical_engineering": 40.7, "mmlu_elementary_mathematics": 26.2, "mmlu_formal_logic": 27.0, "mmlu_global_facts": 31.0, "mmlu_high_school_biology": 48.4, "mmlu_high_school_chemistry": 40.9, "mmlu_high_school_computer_science": 42.0, "mmlu_high_school_european_history": 41.8, "mmlu_high_school_geography": 50.5, "mmlu_high_school_government_and_politics": 59.1, "mmlu_high_school_macroeconomics": 41.0, "mmlu_high_school_mathematics": 25.9, "mmlu_high_school_microeconomics": 38.7, "mmlu_high_school_physics": 25.2, "mmlu_high_school_psychology": 57.2, "mmlu_high_school_statistics": 35.2, "mmlu_high_school_us_history": 44.1, "mmlu_high_school_world_history": 39.7, "mmlu_human_aging": 44.4, "mmlu_human_sexuality": 45.8, "mmlu_international_law": 61.2, "mmlu_jurisprudence": 41.7, "mmlu_logical_fallacies": 41.1, "mmlu_machine_learning": 39.3, "mmlu_management": 56.3, "mmlu_marketing": 60.7, "mmlu_medical_genetics": 43.0, "mmlu_miscellaneous": 54.7, "mmlu_moral_disputes": 43.9, "mmlu_moral_scenarios": 23.7, "mmlu_nutrition": 46.7, "mmlu_philosophy": 41.2, "mmlu_prehistory": 46.3, "mmlu_pro": 13.7, "mmlu_professional_accounting": 34.0, "mmlu_professional_law": 34.7, "mmlu_professional_medicine": 35.3, "mmlu_professional_psychology": 37.4, "mmlu_public_relations": 47.3, "mmlu_security_studies": 46.1, "mmlu_sociology": 42.8, "mmlu_us_foreign_policy": 57.0, "mmlu_virology": 44.6, "mmlu_world_religions": 54.4, "musr": 39.8, "truthfulqa": 33.1, "winogrande": 66.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "gemma 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/gemma-2b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2024-07"}, {"benchmark_scores": {"ai2d": 84.2, "arena_elo_coding": 1240.0, "arena_elo_hard_prompts": 1331.5, "arena_elo_math": 1220.0, "arena_elo_overall": 1230.0, "arena_elo_style_control": 1341.4, "artificial_analysis_quality_index": 65.0, "artificial_analysis_speed_index": 88.0, "bbh": 48.7, "chartqa": 75.7, "docvqa": 87.1, "gpqa_diamond": 45.5, "gsm8k": 94.4, "humaneval": 72.0, "ifeval": 73.0, "live_code_bench": 24.6, "math_500": 65.3, "mathvista": 62.9, "mmlu_pro": 58.2, "mmmu": 59.6, "musr": 20.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2"}, "context_window": 32768, "cost_input": 0.0, "display_name": "Gemma 3 12B", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "google/gemma-3-12b-it", "model_subtypes": ["llm-code"], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 62.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.02, "display_name": "gemma 3 1B it", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "google/gemma-3-1b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 62.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.02, "display_name": "gemma 3 1B pt", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "google/gemma-3-1b-pt", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "gemma 3 270M it", "fits": {"amd_rx_7900_xt": 4148.1, "amd_rx_7900_xtx": 4977.8, "amd_rx_9070_xt": 3344.4, "apple_m1_max": 2074.1, "apple_m2_max": 2074.1, "apple_m2_ultra": 4148.1, "apple_m3_max": 2074.1, "apple_m3_ultra": 4246.7, "apple_m4": 622.2, "apple_m4_max": 2831.1, "apple_m4_pro": 1415.6, "nvidia_rtx_3060_12gb": 1866.7, "nvidia_rtx_3090": 4853.3, "nvidia_rtx_4060_ti_16gb": 1493.3, "nvidia_rtx_4070_ti_super": 3484.4, "nvidia_rtx_4080_super": 3816.3, "nvidia_rtx_4090": 5226.7, "nvidia_rtx_5080": 4977.8, "nvidia_rtx_5090": 9291.9}, "model_id": "google/gemma-3-270m-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "gemma 3 270M", "fits": {"amd_rx_7900_xt": 4148.1, "amd_rx_7900_xtx": 4977.8, "amd_rx_9070_xt": 3344.4, "apple_m1_max": 2074.1, "apple_m2_max": 2074.1, "apple_m2_ultra": 4148.1, "apple_m3_max": 2074.1, "apple_m3_ultra": 4246.7, "apple_m4": 622.2, "apple_m4_max": 2831.1, "apple_m4_pro": 1415.6, "nvidia_rtx_3060_12gb": 1866.7, "nvidia_rtx_3090": 4853.3, "nvidia_rtx_4060_ti_16gb": 1493.3, "nvidia_rtx_4070_ti_super": 3484.4, "nvidia_rtx_4080_super": 3816.3, "nvidia_rtx_4090": 5226.7, "nvidia_rtx_5080": 4977.8, "nvidia_rtx_5090": 9291.9}, "model_id": "google/gemma-3-270m", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"ai2d": 84.5, "aider_polyglot": 4.9, "arena_elo_coding": 1270.0, "arena_elo_hard_prompts": 1364.9, "arena_elo_math": 1250.0, "arena_elo_overall": 1260.0, "arena_elo_style_control": 1365.1, "arena_elo_vision": 1157.3, "artificial_analysis_quality_index": 70.0, "artificial_analysis_speed_index": 82.0, "bbh": 56.2, "chartqa": 72.5, "docvqa": 82.1, "gpqa_diamond": 58.2, "gsm8k": 95.9, "humaneval": 79.5, "ifeval": 79.0, "live_code_bench": 29.7, "math_500": 77.8, "mathvista": 48.5, "medmcqa": 62.6, "medqa": 74.9, "mmlu_pro": 67.5, "mmmu": 55.2, "musr": 25.2, "pubmedqa": 73.4}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.0, "display_name": "Gemma 3 27B", "fits": {"amd_rx_7900_xt": 41.5, "amd_rx_7900_xtx": 49.8, "apple_m1_max": 20.7, "apple_m2_max": 20.7, "apple_m2_ultra": 41.5, "apple_m3_max": 20.7, "apple_m3_ultra": 42.5, "apple_m4": 6.2, "apple_m4_max": 28.3, "apple_m4_pro": 14.2, "nvidia_rtx_3090": 48.5, "nvidia_rtx_4090": 52.3, "nvidia_rtx_5090": 92.9}, "model_id": "google/gemma-3-27b-it", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 74.8, "arena_elo_coding": 1190.0, "arena_elo_hard_prompts": 1283.4, "arena_elo_math": 1170.0, "arena_elo_overall": 1180.0, "arena_elo_style_control": 1302.8, "chartqa": 68.8, "docvqa": 75.8, "gpqa_diamond": 33.2, "gsm8k": 89.2, "humaneval": 58.5, "ifeval": 62.0, "live_code_bench": 12.6, "math_500": 48.5, "mathvista": 50.0, "mmlu_pro": 43.1, "mmmu": 48.8}, "capability_tiers": {}, "context_window": 32768, "cost_input": 0.0, "display_name": "Gemma 3 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/gemma-3-4b-it", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 74.8, "arena_elo_coding": 1190.0, "arena_elo_math": 1170.0, "arena_elo_overall": 1180.0, "chartqa": 68.8, "docvqa": 75.8, "gpqa_diamond": 33.2, "gsm8k": 89.2, "humaneval": 58.5, "ifeval": 62.0, "math_500": 48.5, "mathvista": 50.0, "mmlu_pro": 43.1, "mmmu": 48.8}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.04, "display_name": "gemma 3 4B pt", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/gemma-3-4b-pt", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.0, "display_name": "Gemma 3n 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/gemma-3n-e2b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"mgsm": 67.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.0, "display_name": "Gemma 3n 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/gemma-3n-e4b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_hard_prompts": 1461.1, "arena_elo_math": 1270.0, "arena_elo_overall": 1280.0, "arena_elo_style_control": 1437.9, "gpqa_diamond": 65.2, "humaneval": 82.5, "ifeval": 81.5, "math_500": 85.1, "mmlu_pro": 72.3, "swe_bench_verified": 28.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.2, "display_name": "gemma 4 26B A4B it", "fits": {"amd_rx_7900_xt": 43.1, "amd_rx_7900_xtx": 51.7, "apple_m1_max": 21.5, "apple_m2_max": 21.5, "apple_m2_ultra": 43.1, "apple_m3_max": 21.5, "apple_m3_ultra": 44.1, "apple_m4": 6.5, "apple_m4_max": 29.4, "apple_m4_pro": 14.7, "nvidia_rtx_3090": 50.4, "nvidia_rtx_4090": 54.3, "nvidia_rtx_5090": 96.5}, "model_id": "google/gemma-4-26b-a4b-it", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 52.1, "arena_elo_coding": 1290.0, "arena_elo_math": 1270.0, "arena_elo_overall": 1280.0, "bbh": 60.1, "gpqa_diamond": 65.2, "humaneval": 82.5, "ifeval": 81.5, "math_500": 85.1, "mmlu_astronomy": 72.2, "mmlu_biology": 80.8, "mmlu_business_ethics": 73.5, "mmlu_chemistry": 73.2, "mmlu_clinical_knowledge": 79.2, "mmlu_computer_science": 81.2, "mmlu_jurisprudence": 70.2, "mmlu_physics": 77.5, "mmlu_pro": 72.3, "mmlu_professional_accounting": 64.5, "mmlu_professional_law": 69.2, "multipl_e": 75.4, "multipl_e_cpp": 74.8, "multipl_e_csharp": 72.5, "multipl_e_go": 71.5, "multipl_e_java": 79.2, "multipl_e_javascript": 77.5, "multipl_e_julia": 52.5, "multipl_e_kotlin": 62.8, "multipl_e_lua": 45.2, "multipl_e_perl": 40.5, "multipl_e_php": 69.8, "multipl_e_python": 82.5, "multipl_e_r": 48.2, "multipl_e_ruby": 55.5, "multipl_e_rust": 65.2, "multipl_e_scala": 50.2, "multipl_e_swift": 55.5, "multipl_e_typescript": 76.8, "musr": 28.7, "swe_bench_verified": 28.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.2, "display_name": "Gemma 4 26B", "fits": {"amd_rx_7900_xt": 43.1, "amd_rx_7900_xtx": 51.7, "apple_m1_max": 21.5, "apple_m2_max": 21.5, "apple_m2_ultra": 43.1, "apple_m3_max": 21.5, "apple_m3_ultra": 44.1, "apple_m4": 6.5, "apple_m4_max": 29.4, "apple_m4_pro": 14.7, "nvidia_rtx_3090": 50.4, "nvidia_rtx_4090": 54.3, "nvidia_rtx_5090": 96.5}, "model_id": "google/gemma-4-26b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 54.8, "arena_elo_coding": 1300.0, "arena_elo_math": 1280.0, "arena_elo_overall": 1290.0, "artificial_analysis_quality_index": 73.0, "artificial_analysis_speed_index": 84.0, "bbh": 62.3, "chartqa": 78.5, "docvqa": 86.8, "gpqa_diamond": 67.8, "humaneval": 84.1, "ifeval": 83.0, "math_500": 87.3, "mathvista": 54.5, "mmlu_astronomy": 73.5, "mmlu_biology": 81.5, "mmlu_business_ethics": 74.2, "mmlu_chemistry": 74.5, "mmlu_clinical_knowledge": 80.2, "mmlu_computer_science": 82.5, "mmlu_jurisprudence": 71.5, "mmlu_physics": 78.8, "mmlu_pro": 74.1, "mmlu_professional_accounting": 65.2, "mmlu_professional_law": 70.2, "mmmu": 60.2, "multipl_e": 77.3, "multipl_e_cpp": 76.5, "multipl_e_csharp": 74.2, "multipl_e_go": 73.2, "multipl_e_java": 81.2, "multipl_e_javascript": 79.5, "multipl_e_julia": 54.2, "multipl_e_kotlin": 65.2, "multipl_e_lua": 47.5, "multipl_e_perl": 42.5, "multipl_e_php": 71.5, "multipl_e_python": 84.1, "multipl_e_r": 50.5, "multipl_e_ruby": 57.8, "multipl_e_rust": 67.5, "multipl_e_scala": 52.5, "multipl_e_swift": 57.8, "multipl_e_typescript": 78.8, "musr": 30.4, "swe_bench_verified": 30.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.2, "display_name": "gemma 4 31B it", "fits": {"amd_rx_7900_xtx": 43.4, "apple_m1_max": 18.1, "apple_m2_max": 18.1, "apple_m2_ultra": 36.1, "apple_m3_max": 18.1, "apple_m3_ultra": 37.0, "apple_m4": 5.4, "apple_m4_max": 24.7, "apple_m4_pro": 12.3, "nvidia_rtx_3090": 42.3, "nvidia_rtx_4090": 45.5, "nvidia_rtx_5090": 80.9}, "model_id": "google/gemma-4-31b-it", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 54.8, "arena_elo_coding": 1300.0, "arena_elo_hard_prompts": 1474.2, "arena_elo_math": 1280.0, "arena_elo_overall": 1290.0, "arena_elo_style_control": 1451.2, "artificial_analysis_quality_index": 73.0, "artificial_analysis_speed_index": 84.0, "bbh": 62.3, "chartqa": 78.5, "docvqa": 86.8, "gpqa_diamond": 67.8, "humaneval": 84.1, "ifeval": 83.0, "math_500": 87.3, "mathvista": 54.5, "mmlu_astronomy": 73.5, "mmlu_biology": 81.5, "mmlu_business_ethics": 74.2, "mmlu_chemistry": 74.5, "mmlu_clinical_knowledge": 80.2, "mmlu_computer_science": 82.5, "mmlu_jurisprudence": 71.5, "mmlu_physics": 78.8, "mmlu_pro": 74.1, "mmlu_professional_accounting": 65.2, "mmlu_professional_law": 70.2, "mmmu": 60.2, "multipl_e": 77.3, "multipl_e_cpp": 76.5, "multipl_e_csharp": 74.2, "multipl_e_go": 73.2, "multipl_e_java": 81.2, "multipl_e_javascript": 79.5, "multipl_e_julia": 54.2, "multipl_e_kotlin": 65.2, "multipl_e_lua": 47.5, "multipl_e_perl": 42.5, "multipl_e_php": 71.5, "multipl_e_python": 84.1, "multipl_e_r": 50.5, "multipl_e_ruby": 57.8, "multipl_e_rust": 67.5, "multipl_e_scala": 52.5, "multipl_e_swift": 57.8, "multipl_e_typescript": 78.8, "musr": 30.4, "swe_bench_verified": 30.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.2, "display_name": "Gemma 4 31B", "fits": {"amd_rx_7900_xtx": 43.4, "apple_m1_max": 18.1, "apple_m2_max": 18.1, "apple_m2_ultra": 36.1, "apple_m3_max": 18.1, "apple_m3_ultra": 37.0, "apple_m4": 5.4, "apple_m4_max": 24.7, "apple_m4_pro": 12.3, "nvidia_rtx_3090": 42.3, "nvidia_rtx_4090": 45.5, "nvidia_rtx_5090": 80.9}, "model_id": "google/gemma-4-31b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.03, "display_name": "gemma 4 E2B it", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/gemma-4-e2b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.05, "display_name": "gemma 4 E4B it", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/gemma-4-e4b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 61.1, "bbh": 36.5, "gpqa_diamond": 28.4, "gsm8k": 52.8, "hellaswag": 82.5, "ifeval": 38.7, "math_500": 2.9, "mmlu_abstract_algebra": 31.0, "mmlu_anatomy": 57.8, "mmlu_astronomy": 73.0, "mmlu_business_ethics": 64.0, "mmlu_clinical_knowledge": 68.7, "mmlu_college_biology": 77.1, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 55.0, "mmlu_college_mathematics": 36.0, "mmlu_college_medicine": 69.9, "mmlu_college_physics": 40.2, "mmlu_computer_security": 73.0, "mmlu_conceptual_physics": 62.1, "mmlu_econometrics": 48.2, "mmlu_electrical_engineering": 62.8, "mmlu_elementary_mathematics": 50.3, "mmlu_formal_logic": 50.0, "mmlu_global_facts": 43.0, "mmlu_high_school_biology": 80.3, "mmlu_high_school_chemistry": 52.2, "mmlu_high_school_computer_science": 67.0, "mmlu_high_school_european_history": 76.4, "mmlu_high_school_geography": 82.3, "mmlu_high_school_government_and_politics": 88.6, "mmlu_high_school_macroeconomics": 64.6, "mmlu_high_school_mathematics": 43.0, "mmlu_high_school_microeconomics": 66.8, "mmlu_high_school_physics": 41.7, "mmlu_high_school_psychology": 83.1, "mmlu_high_school_statistics": 57.9, "mmlu_high_school_us_history": 83.3, "mmlu_high_school_world_history": 85.2, "mmlu_human_aging": 71.7, "mmlu_human_sexuality": 72.5, "mmlu_international_law": 84.3, "mmlu_jurisprudence": 77.8, "mmlu_logical_fallacies": 74.8, "mmlu_machine_learning": 51.8, "mmlu_management": 84.5, "mmlu_marketing": 89.7, "mmlu_medical_genetics": 72.0, "mmlu_miscellaneous": 83.8, "mmlu_moral_disputes": 71.7, "mmlu_moral_scenarios": 40.3, "mmlu_nutrition": 76.8, "mmlu_philosophy": 72.3, "mmlu_prehistory": 73.8, "mmlu_pro": 16.9, "mmlu_professional_accounting": 49.3, "mmlu_professional_law": 48.1, "mmlu_professional_medicine": 63.2, "mmlu_professional_psychology": 68.8, "mmlu_public_relations": 70.9, "mmlu_security_studies": 74.7, "mmlu_sociology": 85.6, "mmlu_us_foreign_policy": 88.0, "mmlu_virology": 53.6, "mmlu_world_religions": 81.9, "musr": 42.7, "truthfulqa": 44.9, "winogrande": 78.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "gemma 7B it", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "google/gemma-7b-it", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "madlad400 3B mt", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "google/madlad400-3b-mt", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"medmcqa": 55.7, "medqa": 64.4, "pubmedqa": 73.4}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "medgemma 1.5 4B it", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/medgemma-1-5-4b-it", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"medmcqa": 74.2, "medqa": 88.5, "pubmedqa": 76.8}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "medgemma 27B it", "fits": {"amd_rx_7900_xt": 41.5, "amd_rx_7900_xtx": 49.8, "apple_m1_max": 20.7, "apple_m2_max": 20.7, "apple_m2_ultra": 41.5, "apple_m3_max": 20.7, "apple_m3_ultra": 42.5, "apple_m4": 6.2, "apple_m4_max": 28.3, "apple_m4_pro": 14.2, "nvidia_rtx_3090": 48.5, "nvidia_rtx_4090": 52.3, "nvidia_rtx_5090": 92.9}, "model_id": "google/medgemma-27b-it", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {"medmcqa": 55.7, "medqa": 72.1, "pubmedqa": 73.4}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "medgemma 4B it", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/medgemma-4b-it", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "mobilebert uncased", "fits": {"amd_rx_7900_xt": 44800.0, "amd_rx_7900_xtx": 53760.0, "amd_rx_9070_xt": 36120.0, "apple_m1_max": 22400.0, "apple_m2_max": 22400.0, "apple_m2_ultra": 44800.0, "apple_m3_max": 22400.0, "apple_m3_ultra": 45864.0, "apple_m4": 6720.0, "apple_m4_max": 30576.0, "apple_m4_pro": 15288.0, "nvidia_rtx_3060_12gb": 20160.0, "nvidia_rtx_3090": 52416.0, "nvidia_rtx_4060_ti_16gb": 16128.0, "nvidia_rtx_4070_ti_super": 37632.0, "nvidia_rtx_4080_super": 41216.0, "nvidia_rtx_4090": 56448.0, "nvidia_rtx_5080": 53760.0, "nvidia_rtx_5090": 100352.0}, "model_id": "google/mobilebert-uncased", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "mobilenet v2 1.0 224", "fits": {"amd_rx_7900_xt": 329411.8, "amd_rx_7900_xtx": 395294.1, "amd_rx_9070_xt": 265588.2, "apple_m1_max": 164705.9, "apple_m2_max": 164705.9, "apple_m2_ultra": 329411.8, "apple_m3_max": 164705.9, "apple_m3_ultra": 337235.3, "apple_m4": 49411.8, "apple_m4_max": 224823.5, "apple_m4_pro": 112411.8, "nvidia_rtx_3060_12gb": 148235.3, "nvidia_rtx_3090": 385411.8, "nvidia_rtx_4060_ti_16gb": 118588.2, "nvidia_rtx_4070_ti_super": 276705.9, "nvidia_rtx_4080_super": 303058.8, "nvidia_rtx_4090": 415058.8, "nvidia_rtx_5080": 395294.1, "nvidia_rtx_5090": 737882.4}, "model_id": "google/mobilenet-v2-1-0-224", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"bbh": 28.8, "gpqa_diamond": 23.9, "ifeval": 16.5, "math_500": 0.9, "mmlu_pro": 10.7, "musr": 36.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "mt5 base", "fits": {"amd_rx_7900_xt": 1931.0, "amd_rx_7900_xtx": 2317.2, "amd_rx_9070_xt": 1556.9, "apple_m1_max": 965.5, "apple_m2_max": 965.5, "apple_m2_ultra": 1931.0, "apple_m3_max": 965.5, "apple_m3_ultra": 1976.9, "apple_m4": 289.7, "apple_m4_max": 1317.9, "apple_m4_pro": 659.0, "nvidia_rtx_3060_12gb": 869.0, "nvidia_rtx_3090": 2259.3, "nvidia_rtx_4060_ti_16gb": 695.2, "nvidia_rtx_4070_ti_super": 1622.1, "nvidia_rtx_4080_super": 1776.6, "nvidia_rtx_4090": 2433.1, "nvidia_rtx_5080": 2317.2, "nvidia_rtx_5090": 4325.5}, "model_id": "google/mt5-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "mt5 large", "fits": {"amd_rx_7900_xt": 933.3, "amd_rx_7900_xtx": 1120.0, "amd_rx_9070_xt": 752.5, "apple_m1_max": 466.7, "apple_m2_max": 466.7, "apple_m2_ultra": 933.3, "apple_m3_max": 466.7, "apple_m3_ultra": 955.5, "apple_m4": 140.0, "apple_m4_max": 637.0, "apple_m4_pro": 318.5, "nvidia_rtx_3060_12gb": 420.0, "nvidia_rtx_3090": 1092.0, "nvidia_rtx_4060_ti_16gb": 336.0, "nvidia_rtx_4070_ti_super": 784.0, "nvidia_rtx_4080_super": 858.7, "nvidia_rtx_4090": 1176.0, "nvidia_rtx_5080": 1120.0, "nvidia_rtx_5090": 2090.7}, "model_id": "google/mt5-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"bbh": 27.7, "gpqa_diamond": 24.2, "ifeval": 17.2, "math_500": 0.0, "mmlu_pro": 11.2, "musr": 38.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "mt5 small", "fits": {"amd_rx_7900_xt": 3733.3, "amd_rx_7900_xtx": 4480.0, "amd_rx_9070_xt": 3010.0, "apple_m1_max": 1866.7, "apple_m2_max": 1866.7, "apple_m2_ultra": 3733.3, "apple_m3_max": 1866.7, "apple_m3_ultra": 3822.0, "apple_m4": 560.0, "apple_m4_max": 2548.0, "apple_m4_pro": 1274.0, "nvidia_rtx_3060_12gb": 1680.0, "nvidia_rtx_3090": 4368.0, "nvidia_rtx_4060_ti_16gb": 1344.0, "nvidia_rtx_4070_ti_super": 3136.0, "nvidia_rtx_4080_super": 3434.7, "nvidia_rtx_4090": 4704.0, "nvidia_rtx_5080": 4480.0, "nvidia_rtx_5090": 8362.7}, "model_id": "google/mt5-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "owlv2 base patch16 ensemble", "fits": {"amd_rx_7900_xt": 8615.4, "amd_rx_7900_xtx": 10338.5, "amd_rx_9070_xt": 6946.2, "apple_m1_max": 4307.7, "apple_m2_max": 4307.7, "apple_m2_ultra": 8615.4, "apple_m3_max": 4307.7, "apple_m3_ultra": 8820.0, "apple_m4": 1292.3, "apple_m4_max": 5880.0, "apple_m4_pro": 2940.0, "nvidia_rtx_3060_12gb": 3876.9, "nvidia_rtx_3090": 10080.0, "nvidia_rtx_4060_ti_16gb": 3101.5, "nvidia_rtx_4070_ti_super": 7236.9, "nvidia_rtx_4080_super": 7926.2, "nvidia_rtx_4090": 10855.4, "nvidia_rtx_5080": 10338.5, "nvidia_rtx_5090": 19298.5}, "model_id": "google/owlv2-base-patch16-ensemble", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "owlv2 base patch16", "fits": {"amd_rx_7900_xt": 8615.4, "amd_rx_7900_xtx": 10338.5, "amd_rx_9070_xt": 6946.2, "apple_m1_max": 4307.7, "apple_m2_max": 4307.7, "apple_m2_ultra": 8615.4, "apple_m3_max": 4307.7, "apple_m3_ultra": 8820.0, "apple_m4": 1292.3, "apple_m4_max": 5880.0, "apple_m4_pro": 2940.0, "nvidia_rtx_3060_12gb": 3876.9, "nvidia_rtx_3090": 10080.0, "nvidia_rtx_4060_ti_16gb": 3101.5, "nvidia_rtx_4070_ti_super": 7236.9, "nvidia_rtx_4080_super": 7926.2, "nvidia_rtx_4090": 10855.4, "nvidia_rtx_5080": 10338.5, "nvidia_rtx_5090": 19298.5}, "model_id": "google/owlv2-base-patch16", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "owlv2 large patch14 ensemble", "fits": {"amd_rx_7900_xt": 2616.8, "amd_rx_7900_xtx": 3140.2, "amd_rx_9070_xt": 2109.8, "apple_m1_max": 1308.4, "apple_m2_max": 1308.4, "apple_m2_ultra": 2616.8, "apple_m3_max": 1308.4, "apple_m3_ultra": 2679.0, "apple_m4": 392.5, "apple_m4_max": 1786.0, "apple_m4_pro": 893.0, "nvidia_rtx_3060_12gb": 1177.6, "nvidia_rtx_3090": 3061.7, "nvidia_rtx_4060_ti_16gb": 942.1, "nvidia_rtx_4070_ti_super": 2198.1, "nvidia_rtx_4080_super": 2407.5, "nvidia_rtx_4090": 3297.2, "nvidia_rtx_5080": 3140.2, "nvidia_rtx_5090": 5861.7}, "model_id": "google/owlv2-large-patch14-ensemble", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "owlvit base patch32", "fits": {"amd_rx_7900_xt": 7368.4, "amd_rx_7900_xtx": 8842.1, "amd_rx_9070_xt": 5940.8, "apple_m1_max": 3684.2, "apple_m2_max": 3684.2, "apple_m2_ultra": 7368.4, "apple_m3_max": 3684.2, "apple_m3_ultra": 7543.4, "apple_m4": 1105.3, "apple_m4_max": 5028.9, "apple_m4_pro": 2514.5, "nvidia_rtx_3060_12gb": 3315.8, "nvidia_rtx_3090": 8621.1, "nvidia_rtx_4060_ti_16gb": 2652.6, "nvidia_rtx_4070_ti_super": 6189.5, "nvidia_rtx_4080_super": 6778.9, "nvidia_rtx_4090": 9284.2, "nvidia_rtx_5080": 8842.1, "nvidia_rtx_5090": 16505.3}, "model_id": "google/owlvit-base-patch32", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "paligemma 3B mix 224", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "google/paligemma-3b-mix-224", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "paligemma 3B pt 224", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "google/paligemma-3b-pt-224", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "pegasus xsum", "fits": {"amd_rx_7900_xt": 1971.8, "amd_rx_7900_xtx": 2366.2, "amd_rx_9070_xt": 1589.8, "apple_m1_max": 985.9, "apple_m2_max": 985.9, "apple_m2_ultra": 1971.8, "apple_m3_max": 985.9, "apple_m3_ultra": 2018.7, "apple_m4": 295.8, "apple_m4_max": 1345.8, "apple_m4_pro": 672.9, "nvidia_rtx_3060_12gb": 887.3, "nvidia_rtx_3090": 2307.0, "nvidia_rtx_4060_ti_16gb": 709.9, "nvidia_rtx_4070_ti_super": 1656.3, "nvidia_rtx_4080_super": 1814.1, "nvidia_rtx_4090": 2484.5, "nvidia_rtx_5080": 2366.2, "nvidia_rtx_5090": 4416.9}, "model_id": "google/pegasus-xsum", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 65536, "cost_input": null, "display_name": "reformer crime and punishment", "fits": {"amd_rx_7900_xt": 7516.8, "amd_rx_7900_xtx": 9020.1, "amd_rx_9070_xt": 6060.4, "apple_m1_max": 3758.4, "apple_m2_max": 3758.4, "apple_m2_ultra": 7516.8, "apple_m3_max": 3758.4, "apple_m3_ultra": 7695.3, "apple_m4": 1127.5, "apple_m4_max": 5130.2, "apple_m4_pro": 2565.1, "nvidia_rtx_3060_12gb": 3382.6, "nvidia_rtx_3090": 8794.6, "nvidia_rtx_4060_ti_16gb": 2706.0, "nvidia_rtx_4070_ti_super": 6314.1, "nvidia_rtx_4080_super": 6915.4, "nvidia_rtx_4090": 9471.1, "nvidia_rtx_5080": 9020.1, "nvidia_rtx_5090": 16837.6}, "model_id": "google/reformer-crime-and-punishment", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "rembert", "fits": {"amd_rx_7900_xt": 1944.4, "amd_rx_7900_xtx": 2333.3, "amd_rx_9070_xt": 1567.7, "apple_m1_max": 972.2, "apple_m2_max": 972.2, "apple_m2_ultra": 1944.4, "apple_m3_max": 972.2, "apple_m3_ultra": 1990.6, "apple_m4": 291.7, "apple_m4_max": 1327.1, "apple_m4_pro": 663.5, "nvidia_rtx_3060_12gb": 875.0, "nvidia_rtx_3090": 2275.0, "nvidia_rtx_4060_ti_16gb": 700.0, "nvidia_rtx_4070_ti_super": 1633.3, "nvidia_rtx_4080_super": 1788.9, "nvidia_rtx_4090": 2450.0, "nvidia_rtx_5080": 2333.3, "nvidia_rtx_5090": 4355.6}, "model_id": "google/rembert", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "shieldgemma 2 4B it", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/shieldgemma-2-4b-it", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"helm_safety": 94.5, "toxigen": 96.8}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "shieldgemma 27B", "fits": {"amd_rx_7900_xt": 41.5, "amd_rx_7900_xtx": 49.8, "apple_m1_max": 20.7, "apple_m2_max": 20.7, "apple_m2_ultra": 41.5, "apple_m3_max": 20.7, "apple_m3_ultra": 42.5, "apple_m4": 6.2, "apple_m4_max": 28.3, "apple_m4_pro": 14.2, "nvidia_rtx_3090": 48.5, "nvidia_rtx_4090": 52.3, "nvidia_rtx_5090": 92.9}, "model_id": "google/shieldgemma-27b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "shieldgemma 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/shieldgemma-2b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"helm_safety": 93.2, "toxigen": 96.1}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "shieldgemma 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "google/shieldgemma-9b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip base patch16 224", "fits": {"amd_rx_7900_xt": 5517.2, "amd_rx_7900_xtx": 6620.7, "amd_rx_9070_xt": 4448.3, "apple_m1_max": 2758.6, "apple_m2_max": 2758.6, "apple_m2_ultra": 5517.2, "apple_m3_max": 2758.6, "apple_m3_ultra": 5648.3, "apple_m4": 827.6, "apple_m4_max": 3765.5, "apple_m4_pro": 1882.8, "nvidia_rtx_3060_12gb": 2482.8, "nvidia_rtx_3090": 6455.2, "nvidia_rtx_4060_ti_16gb": 1986.2, "nvidia_rtx_4070_ti_super": 4634.5, "nvidia_rtx_4080_super": 5075.9, "nvidia_rtx_4090": 6951.7, "nvidia_rtx_5080": 6620.7, "nvidia_rtx_5090": 12358.6}, "model_id": "google/siglip-base-patch16-224", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip so400m patch14 384", "fits": {"amd_rx_7900_xt": 1275.6, "amd_rx_7900_xtx": 1530.8, "amd_rx_9070_xt": 1028.5, "apple_m1_max": 637.8, "apple_m2_max": 637.8, "apple_m2_ultra": 1275.6, "apple_m3_max": 637.8, "apple_m3_ultra": 1305.9, "apple_m4": 191.3, "apple_m4_max": 870.6, "apple_m4_pro": 435.3, "nvidia_rtx_3060_12gb": 574.0, "nvidia_rtx_3090": 1492.5, "nvidia_rtx_4060_ti_16gb": 459.2, "nvidia_rtx_4070_ti_super": 1071.5, "nvidia_rtx_4080_super": 1173.6, "nvidia_rtx_4090": 1607.3, "nvidia_rtx_5080": 1530.8, "nvidia_rtx_5090": 2857.4}, "model_id": "google/siglip-so400m-patch14-384", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 base patch16 224", "fits": {"amd_rx_7900_xt": 5517.2, "amd_rx_7900_xtx": 6620.7, "amd_rx_9070_xt": 4448.3, "apple_m1_max": 2758.6, "apple_m2_max": 2758.6, "apple_m2_ultra": 5517.2, "apple_m3_max": 2758.6, "apple_m3_ultra": 5648.3, "apple_m4": 827.6, "apple_m4_max": 3765.5, "apple_m4_pro": 1882.8, "nvidia_rtx_3060_12gb": 2482.8, "nvidia_rtx_3090": 6455.2, "nvidia_rtx_4060_ti_16gb": 1986.2, "nvidia_rtx_4070_ti_super": 4634.5, "nvidia_rtx_4080_super": 5075.9, "nvidia_rtx_4090": 6951.7, "nvidia_rtx_5080": 6620.7, "nvidia_rtx_5090": 12358.6}, "model_id": "google/siglip2-base-patch16-224", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 base patch16 256", "fits": {"amd_rx_7900_xt": 5517.2, "amd_rx_7900_xtx": 6620.7, "amd_rx_9070_xt": 4448.3, "apple_m1_max": 2758.6, "apple_m2_max": 2758.6, "apple_m2_ultra": 5517.2, "apple_m3_max": 2758.6, "apple_m3_ultra": 5648.3, "apple_m4": 827.6, "apple_m4_max": 3765.5, "apple_m4_pro": 1882.8, "nvidia_rtx_3060_12gb": 2482.8, "nvidia_rtx_3090": 6455.2, "nvidia_rtx_4060_ti_16gb": 1986.2, "nvidia_rtx_4070_ti_super": 4634.5, "nvidia_rtx_4080_super": 5075.9, "nvidia_rtx_4090": 6951.7, "nvidia_rtx_5080": 6620.7, "nvidia_rtx_5090": 12358.6}, "model_id": "google/siglip2-base-patch16-256", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 base patch16 384", "fits": {"amd_rx_7900_xt": 5517.2, "amd_rx_7900_xtx": 6620.7, "amd_rx_9070_xt": 4448.3, "apple_m1_max": 2758.6, "apple_m2_max": 2758.6, "apple_m2_ultra": 5517.2, "apple_m3_max": 2758.6, "apple_m3_ultra": 5648.3, "apple_m4": 827.6, "apple_m4_max": 3765.5, "apple_m4_pro": 1882.8, "nvidia_rtx_3060_12gb": 2482.8, "nvidia_rtx_3090": 6455.2, "nvidia_rtx_4060_ti_16gb": 1986.2, "nvidia_rtx_4070_ti_super": 4634.5, "nvidia_rtx_4080_super": 5075.9, "nvidia_rtx_4090": 6951.7, "nvidia_rtx_5080": 6620.7, "nvidia_rtx_5090": 12358.6}, "model_id": "google/siglip2-base-patch16-384", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 base patch16 512", "fits": {"amd_rx_7900_xt": 5517.2, "amd_rx_7900_xtx": 6620.7, "amd_rx_9070_xt": 4448.3, "apple_m1_max": 2758.6, "apple_m2_max": 2758.6, "apple_m2_ultra": 5517.2, "apple_m3_max": 2758.6, "apple_m3_ultra": 5648.3, "apple_m4": 827.6, "apple_m4_max": 3765.5, "apple_m4_pro": 1882.8, "nvidia_rtx_3060_12gb": 2482.8, "nvidia_rtx_3090": 6455.2, "nvidia_rtx_4060_ti_16gb": 1986.2, "nvidia_rtx_4070_ti_super": 4634.5, "nvidia_rtx_4080_super": 5075.9, "nvidia_rtx_4090": 6951.7, "nvidia_rtx_5080": 6620.7, "nvidia_rtx_5090": 12358.6}, "model_id": "google/siglip2-base-patch16-512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 base patch16 naflex", "fits": {"amd_rx_7900_xt": 5517.2, "amd_rx_7900_xtx": 6620.7, "amd_rx_9070_xt": 4448.3, "apple_m1_max": 2758.6, "apple_m2_max": 2758.6, "apple_m2_ultra": 5517.2, "apple_m3_max": 2758.6, "apple_m3_ultra": 5648.3, "apple_m4": 827.6, "apple_m4_max": 3765.5, "apple_m4_pro": 1882.8, "nvidia_rtx_3060_12gb": 2482.8, "nvidia_rtx_3090": 6455.2, "nvidia_rtx_4060_ti_16gb": 1986.2, "nvidia_rtx_4070_ti_super": 4634.5, "nvidia_rtx_4080_super": 5075.9, "nvidia_rtx_4090": 6951.7, "nvidia_rtx_5080": 6620.7, "nvidia_rtx_5090": 12358.6}, "model_id": "google/siglip2-base-patch16-naflex", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 giant opt patch16 384", "fits": {"amd_rx_7900_xt": 1018.2, "amd_rx_7900_xtx": 1221.8, "amd_rx_9070_xt": 820.9, "apple_m1_max": 509.1, "apple_m2_max": 509.1, "apple_m2_ultra": 1018.2, "apple_m3_max": 509.1, "apple_m3_ultra": 1042.4, "apple_m4": 152.7, "apple_m4_max": 694.9, "apple_m4_pro": 347.5, "nvidia_rtx_3060_12gb": 458.2, "nvidia_rtx_3090": 1191.3, "nvidia_rtx_4060_ti_16gb": 366.5, "nvidia_rtx_4070_ti_super": 855.3, "nvidia_rtx_4080_super": 936.7, "nvidia_rtx_4090": 1282.9, "nvidia_rtx_5080": 1221.8, "nvidia_rtx_5090": 2280.7}, "model_id": "google/siglip2-giant-opt-patch16-384", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 large patch16 512", "fits": {"amd_rx_7900_xt": 1717.8, "amd_rx_7900_xtx": 2061.3, "amd_rx_9070_xt": 1385.0, "apple_m1_max": 858.9, "apple_m2_max": 858.9, "apple_m2_ultra": 1717.8, "apple_m3_max": 858.9, "apple_m3_ultra": 1758.6, "apple_m4": 257.7, "apple_m4_max": 1172.4, "apple_m4_pro": 586.2, "nvidia_rtx_3060_12gb": 773.0, "nvidia_rtx_3090": 2009.8, "nvidia_rtx_4060_ti_16gb": 618.4, "nvidia_rtx_4070_ti_super": 1442.9, "nvidia_rtx_4080_super": 1580.4, "nvidia_rtx_4090": 2164.4, "nvidia_rtx_5080": 2061.3, "nvidia_rtx_5090": 3847.9}, "model_id": "google/siglip2-large-patch16-512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 so400m patch14 224", "fits": {"amd_rx_7900_xt": 1275.6, "amd_rx_7900_xtx": 1530.8, "amd_rx_9070_xt": 1028.5, "apple_m1_max": 637.8, "apple_m2_max": 637.8, "apple_m2_ultra": 1275.6, "apple_m3_max": 637.8, "apple_m3_ultra": 1305.9, "apple_m4": 191.3, "apple_m4_max": 870.6, "apple_m4_pro": 435.3, "nvidia_rtx_3060_12gb": 574.0, "nvidia_rtx_3090": 1492.5, "nvidia_rtx_4060_ti_16gb": 459.2, "nvidia_rtx_4070_ti_super": 1071.5, "nvidia_rtx_4080_super": 1173.6, "nvidia_rtx_4090": 1607.3, "nvidia_rtx_5080": 1530.8, "nvidia_rtx_5090": 2857.4}, "model_id": "google/siglip2-so400m-patch14-224", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 so400m patch14 384", "fits": {"amd_rx_7900_xt": 1275.6, "amd_rx_7900_xtx": 1530.8, "amd_rx_9070_xt": 1028.5, "apple_m1_max": 637.8, "apple_m2_max": 637.8, "apple_m2_ultra": 1275.6, "apple_m3_max": 637.8, "apple_m3_ultra": 1305.9, "apple_m4": 191.3, "apple_m4_max": 870.6, "apple_m4_pro": 435.3, "nvidia_rtx_3060_12gb": 574.0, "nvidia_rtx_3090": 1492.5, "nvidia_rtx_4060_ti_16gb": 459.2, "nvidia_rtx_4070_ti_super": 1071.5, "nvidia_rtx_4080_super": 1173.6, "nvidia_rtx_4090": 1607.3, "nvidia_rtx_5080": 1530.8, "nvidia_rtx_5090": 2857.4}, "model_id": "google/siglip2-so400m-patch14-384", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 so400m patch16 256", "fits": {"amd_rx_7900_xt": 1275.6, "amd_rx_7900_xtx": 1530.8, "amd_rx_9070_xt": 1028.5, "apple_m1_max": 637.8, "apple_m2_max": 637.8, "apple_m2_ultra": 1275.6, "apple_m3_max": 637.8, "apple_m3_ultra": 1305.9, "apple_m4": 191.3, "apple_m4_max": 870.6, "apple_m4_pro": 435.3, "nvidia_rtx_3060_12gb": 574.0, "nvidia_rtx_3090": 1492.5, "nvidia_rtx_4060_ti_16gb": 459.2, "nvidia_rtx_4070_ti_super": 1071.5, "nvidia_rtx_4080_super": 1173.6, "nvidia_rtx_4090": 1607.3, "nvidia_rtx_5080": 1530.8, "nvidia_rtx_5090": 2857.4}, "model_id": "google/siglip2-so400m-patch16-256", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 so400m patch16 384", "fits": {"amd_rx_7900_xt": 1275.6, "amd_rx_7900_xtx": 1530.8, "amd_rx_9070_xt": 1028.5, "apple_m1_max": 637.8, "apple_m2_max": 637.8, "apple_m2_ultra": 1275.6, "apple_m3_max": 637.8, "apple_m3_ultra": 1305.9, "apple_m4": 191.3, "apple_m4_max": 870.6, "apple_m4_pro": 435.3, "nvidia_rtx_3060_12gb": 574.0, "nvidia_rtx_3090": 1492.5, "nvidia_rtx_4060_ti_16gb": 459.2, "nvidia_rtx_4070_ti_super": 1071.5, "nvidia_rtx_4080_super": 1173.6, "nvidia_rtx_4090": 1607.3, "nvidia_rtx_5080": 1530.8, "nvidia_rtx_5090": 2857.4}, "model_id": "google/siglip2-so400m-patch16-384", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 so400m patch16 512", "fits": {"amd_rx_7900_xt": 1275.6, "amd_rx_7900_xtx": 1530.8, "amd_rx_9070_xt": 1028.5, "apple_m1_max": 637.8, "apple_m2_max": 637.8, "apple_m2_ultra": 1275.6, "apple_m3_max": 637.8, "apple_m3_ultra": 1305.9, "apple_m4": 191.3, "apple_m4_max": 870.6, "apple_m4_pro": 435.3, "nvidia_rtx_3060_12gb": 574.0, "nvidia_rtx_3090": 1492.5, "nvidia_rtx_4060_ti_16gb": 459.2, "nvidia_rtx_4070_ti_super": 1071.5, "nvidia_rtx_4080_super": 1173.6, "nvidia_rtx_4090": 1607.3, "nvidia_rtx_5080": 1530.8, "nvidia_rtx_5090": 2857.4}, "model_id": "google/siglip2-so400m-patch16-512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "siglip2 so400m patch16 naflex", "fits": {"amd_rx_7900_xt": 1275.6, "amd_rx_7900_xtx": 1530.8, "amd_rx_9070_xt": 1028.5, "apple_m1_max": 637.8, "apple_m2_max": 637.8, "apple_m2_ultra": 1275.6, "apple_m3_max": 637.8, "apple_m3_ultra": 1305.9, "apple_m4": 191.3, "apple_m4_max": 870.6, "apple_m4_pro": 435.3, "nvidia_rtx_3060_12gb": 574.0, "nvidia_rtx_3090": 1492.5, "nvidia_rtx_4060_ti_16gb": 459.2, "nvidia_rtx_4070_ti_super": 1071.5, "nvidia_rtx_4080_super": 1173.6, "nvidia_rtx_4090": 1607.3, "nvidia_rtx_5080": 1530.8, "nvidia_rtx_5090": 2857.4}, "model_id": "google/siglip2-so400m-patch16-naflex", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "t5 v1 1 large", "fits": {"amd_rx_7900_xt": 1454.5, "amd_rx_7900_xtx": 1745.5, "amd_rx_9070_xt": 1172.7, "apple_m1_max": 727.3, "apple_m2_max": 727.3, "apple_m2_ultra": 1454.5, "apple_m3_max": 727.3, "apple_m3_ultra": 1489.1, "apple_m4": 218.2, "apple_m4_max": 992.7, "apple_m4_pro": 496.4, "nvidia_rtx_3060_12gb": 654.5, "nvidia_rtx_3090": 1701.8, "nvidia_rtx_4060_ti_16gb": 523.6, "nvidia_rtx_4070_ti_super": 1221.8, "nvidia_rtx_4080_super": 1338.2, "nvidia_rtx_4090": 1832.7, "nvidia_rtx_5080": 1745.5, "nvidia_rtx_5090": 3258.2}, "model_id": "google/t5-v1-1-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "t5 v1 1 xxl", "fits": {"amd_rx_7900_xt": 101.8, "amd_rx_7900_xtx": 122.2, "amd_rx_9070_xt": 82.1, "apple_m1_max": 50.9, "apple_m2_max": 50.9, "apple_m2_ultra": 101.8, "apple_m3_max": 50.9, "apple_m3_ultra": 104.2, "apple_m4": 15.3, "apple_m4_max": 69.5, "apple_m4_pro": 34.7, "nvidia_rtx_3060_12gb": 45.8, "nvidia_rtx_3090": 119.1, "nvidia_rtx_4060_ti_16gb": 36.7, "nvidia_rtx_4070_ti_super": 85.5, "nvidia_rtx_4080_super": 93.7, "nvidia_rtx_4090": 128.3, "nvidia_rtx_5080": 122.2, "nvidia_rtx_5090": 228.1}, "model_id": "google/t5-v1-1-xxl", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "t5gemma 2 1B 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "google/t5gemma-2-1b-1b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "t5gemma s s prefixlm", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "google/t5gemma-s-s-prefixlm", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "tapas large finetuned sqa", "fits": {"amd_rx_7900_xt": 3294.1, "amd_rx_7900_xtx": 3952.9, "amd_rx_9070_xt": 2655.9, "apple_m1_max": 1647.1, "apple_m2_max": 1647.1, "apple_m2_ultra": 3294.1, "apple_m3_max": 1647.1, "apple_m3_ultra": 3372.4, "apple_m4": 494.1, "apple_m4_max": 2248.2, "apple_m4_pro": 1124.1, "nvidia_rtx_3060_12gb": 1482.4, "nvidia_rtx_3090": 3854.1, "nvidia_rtx_4060_ti_16gb": 1185.9, "nvidia_rtx_4070_ti_super": 2767.1, "nvidia_rtx_4080_super": 3030.6, "nvidia_rtx_4090": 4150.6, "nvidia_rtx_5080": 3952.9, "nvidia_rtx_5090": 7378.8}, "model_id": "google/tapas-large-finetuned-sqa", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "translategemma 12B it", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "google/translategemma-12b-it", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "translategemma 4B it", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "google/translategemma-4b-it", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "umt5 xxl", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "google/umt5-xxl", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "vit base patch16 224 in21k", "fits": {"amd_rx_7900_xt": 13023.3, "amd_rx_7900_xtx": 15627.9, "amd_rx_9070_xt": 10500.0, "apple_m1_max": 6511.6, "apple_m2_max": 6511.6, "apple_m2_ultra": 13023.3, "apple_m3_max": 6511.6, "apple_m3_ultra": 13332.6, "apple_m4": 1953.5, "apple_m4_max": 8888.4, "apple_m4_pro": 4444.2, "nvidia_rtx_3060_12gb": 5860.5, "nvidia_rtx_3090": 15237.2, "nvidia_rtx_4060_ti_16gb": 4688.4, "nvidia_rtx_4070_ti_super": 10939.5, "nvidia_rtx_4080_super": 11981.4, "nvidia_rtx_4090": 16409.3, "nvidia_rtx_5080": 15627.9, "nvidia_rtx_5090": 29172.1}, "model_id": "google/vit-base-patch16-224-in21k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "vit base patch16 224", "fits": {"amd_rx_7900_xt": 13023.3, "amd_rx_7900_xtx": 15627.9, "amd_rx_9070_xt": 10500.0, "apple_m1_max": 6511.6, "apple_m2_max": 6511.6, "apple_m2_ultra": 13023.3, "apple_m3_max": 6511.6, "apple_m3_ultra": 13332.6, "apple_m4": 1953.5, "apple_m4_max": 8888.4, "apple_m4_pro": 4444.2, "nvidia_rtx_3060_12gb": 5860.5, "nvidia_rtx_3090": 15237.2, "nvidia_rtx_4060_ti_16gb": 4688.4, "nvidia_rtx_4070_ti_super": 10939.5, "nvidia_rtx_4080_super": 11981.4, "nvidia_rtx_4090": 16409.3, "nvidia_rtx_5080": 15627.9, "nvidia_rtx_5090": 29172.1}, "model_id": "google/vit-base-patch16-224", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "vit large patch16 384", "fits": {"amd_rx_7900_xt": 3684.2, "amd_rx_7900_xtx": 4421.1, "amd_rx_9070_xt": 2970.4, "apple_m1_max": 1842.1, "apple_m2_max": 1842.1, "apple_m2_ultra": 3684.2, "apple_m3_max": 1842.1, "apple_m3_ultra": 3771.7, "apple_m4": 552.6, "apple_m4_max": 2514.5, "apple_m4_pro": 1257.2, "nvidia_rtx_3060_12gb": 1657.9, "nvidia_rtx_3090": 4310.5, "nvidia_rtx_4060_ti_16gb": 1326.3, "nvidia_rtx_4070_ti_super": 3094.7, "nvidia_rtx_4080_super": 3389.5, "nvidia_rtx_4090": 4642.1, "nvidia_rtx_5080": 4421.1, "nvidia_rtx_5090": 8252.6}, "model_id": "google/vit-large-patch16-384", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Google DeepMind", "scores_as_of": null}, {"benchmark_scores": {"bbh": 32.2, "gpqa_diamond": 24.7, "ifeval": 24.0, "math_500": 2.6, "mmlu_pro": 11.5, "musr": 33.7}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "granite 3.0 1B a400m base", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "ibm/granite-3-0-1b-a400m-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "IBM", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 51.9, "gpqa_diamond": 33.2, "ifeval": 53.1, "math_500": 14.2, "mmlu_pro": 34.6, "musr": 39.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 3.0 8B instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "ibm/granite-3-0-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 32.8, "gpqa_diamond": 24.0, "ifeval": 46.9, "math_500": 4.5, "mmlu_pro": 12.2, "musr": 33.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 3.1 1B a400m instruct", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "ibm/granite-3-1-1b-a400m-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 44.1, "gpqa_diamond": 28.9, "ifeval": 62.9, "math_500": 15.3, "mmlu_pro": 28.2, "musr": 36.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 3.1 2B instruct", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "ibm/granite-3-1-2b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 39.5, "gpqa_diamond": 29.8, "ifeval": 72.1, "math_500": 46.8, "mmlu_pro": 43.5, "musr": 13.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 3.1 8B instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "ibm/granite-3-1-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 3.3 2B instruct", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "ibm/granite-3-3-2b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {"gsm8k": 80.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 3.3 8B instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "ibm/granite-3-3-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "granite 3B code base 2K", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "ibm/granite-3b-code-base-2k", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite 4.0 1B speech", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "ibm/granite-4-0-1b-speech", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "ibm/granite-4-0-1b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 350M base", "fits": {"amd_rx_7900_xt": 3200.0, "amd_rx_7900_xtx": 3840.0, "amd_rx_9070_xt": 2580.0, "apple_m1_max": 1600.0, "apple_m2_max": 1600.0, "apple_m2_ultra": 3200.0, "apple_m3_max": 1600.0, "apple_m3_ultra": 3276.0, "apple_m4": 480.0, "apple_m4_max": 2184.0, "apple_m4_pro": 1092.0, "nvidia_rtx_3060_12gb": 1440.0, "nvidia_rtx_3090": 3744.0, "nvidia_rtx_4060_ti_16gb": 1152.0, "nvidia_rtx_4070_ti_super": 2688.0, "nvidia_rtx_4080_super": 2944.0, "nvidia_rtx_4090": 4032.0, "nvidia_rtx_5080": 3840.0, "nvidia_rtx_5090": 7168.0}, "model_id": "ibm/granite-4-0-350m-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 350M", "fits": {"amd_rx_7900_xt": 3200.0, "amd_rx_7900_xtx": 3840.0, "amd_rx_9070_xt": 2580.0, "apple_m1_max": 1600.0, "apple_m2_max": 1600.0, "apple_m2_ultra": 3200.0, "apple_m3_max": 1600.0, "apple_m3_ultra": 3276.0, "apple_m4": 480.0, "apple_m4_max": 2184.0, "apple_m4_pro": 1092.0, "nvidia_rtx_3060_12gb": 1440.0, "nvidia_rtx_3090": 3744.0, "nvidia_rtx_4060_ti_16gb": 1152.0, "nvidia_rtx_4070_ti_super": 2688.0, "nvidia_rtx_4080_super": 2944.0, "nvidia_rtx_4090": 4032.0, "nvidia_rtx_5080": 3840.0, "nvidia_rtx_5090": 7168.0}, "model_id": "ibm/granite-4-0-350m", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 h 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "ibm/granite-4-0-h-1b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 h 350M", "fits": {"amd_rx_7900_xt": 3200.0, "amd_rx_7900_xtx": 3840.0, "amd_rx_9070_xt": 2580.0, "apple_m1_max": 1600.0, "apple_m2_max": 1600.0, "apple_m2_ultra": 3200.0, "apple_m3_max": 1600.0, "apple_m3_ultra": 3276.0, "apple_m4": 480.0, "apple_m4_max": 2184.0, "apple_m4_pro": 1092.0, "nvidia_rtx_3060_12gb": 1440.0, "nvidia_rtx_3090": 3744.0, "nvidia_rtx_4060_ti_16gb": 1152.0, "nvidia_rtx_4070_ti_super": 2688.0, "nvidia_rtx_4080_super": 2944.0, "nvidia_rtx_4090": 4032.0, "nvidia_rtx_5080": 3840.0, "nvidia_rtx_5090": 7168.0}, "model_id": "ibm/granite-4-0-h-350m", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 h micro", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "ibm/granite-4-0-h-micro", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 h small", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "ibm/granite-4-0-h-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 h tiny", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "ibm/granite-4-0-h-tiny", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 micro base", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "ibm/granite-4-0-micro-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 micro", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "ibm/granite-4-0-micro", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granite 4.0 tiny preview", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "ibm/granite-4-0-tiny-preview", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "granite 8B code instruct 4K GGUF", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "ibm/granite-8b-code-instruct-4k-gguf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "granite docling 258M", "fits": {"amd_rx_7900_xt": 4341.1, "amd_rx_7900_xtx": 5209.3, "amd_rx_9070_xt": 3500.0, "apple_m1_max": 2170.5, "apple_m2_max": 2170.5, "apple_m2_ultra": 4341.1, "apple_m3_max": 2170.5, "apple_m3_ultra": 4444.2, "apple_m4": 651.2, "apple_m4_max": 2962.8, "apple_m4_pro": 1481.4, "nvidia_rtx_3060_12gb": 1953.5, "nvidia_rtx_3090": 5079.1, "nvidia_rtx_4060_ti_16gb": 1562.8, "nvidia_rtx_4070_ti_super": 3646.5, "nvidia_rtx_4080_super": 3993.8, "nvidia_rtx_4090": 5469.8, "nvidia_rtx_5080": 5209.3, "nvidia_rtx_5090": 9724.0}, "model_id": "ibm/granite-docling-258m", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {"miracl": 50.0, "mteb_classification": 64.0, "mteb_clustering": 42.0, "mteb_overall": 58.0, "mteb_pair_classification": 80.5, "mteb_reranking": 52.0, "mteb_retrieval": 54.0, "mteb_sts": 77.5, "mteb_summarization": 27.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite embedding 107M multilingual", "fits": {"amd_rx_7900_xt": 10467.3, "amd_rx_7900_xtx": 12560.7, "amd_rx_9070_xt": 8439.3, "apple_m1_max": 5233.6, "apple_m2_max": 5233.6, "apple_m2_ultra": 10467.3, "apple_m3_max": 5233.6, "apple_m3_ultra": 10715.9, "apple_m4": 1570.1, "apple_m4_max": 7143.9, "apple_m4_pro": 3572.0, "nvidia_rtx_3060_12gb": 4710.3, "nvidia_rtx_3090": 12246.7, "nvidia_rtx_4060_ti_16gb": 3768.2, "nvidia_rtx_4070_ti_super": 8792.5, "nvidia_rtx_4080_super": 9629.9, "nvidia_rtx_4090": 13188.8, "nvidia_rtx_5080": 12560.7, "nvidia_rtx_5090": 23446.7}, "model_id": "ibm/granite-embedding-107m-multilingual", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 48.0, "miracl": 38.0, "mteb_classification": 64.5, "mteb_clustering": 42.0, "mteb_overall": 58.5, "mteb_pair_classification": 81.5, "mteb_reranking": 53.0, "mteb_retrieval": 54.5, "mteb_sts": 78.5, "mteb_summarization": 27.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite embedding 125M english", "fits": {"amd_rx_7900_xt": 8960.0, "amd_rx_7900_xtx": 10752.0, "amd_rx_9070_xt": 7224.0, "apple_m1_max": 4480.0, "apple_m2_max": 4480.0, "apple_m2_ultra": 8960.0, "apple_m3_max": 4480.0, "apple_m3_ultra": 9172.8, "apple_m4": 1344.0, "apple_m4_max": 6115.2, "apple_m4_pro": 3057.6, "nvidia_rtx_3060_12gb": 4032.0, "nvidia_rtx_3090": 10483.2, "nvidia_rtx_4060_ti_16gb": 3225.6, "nvidia_rtx_4070_ti_super": 7526.4, "nvidia_rtx_4080_super": 8243.2, "nvidia_rtx_4090": 11289.6, "nvidia_rtx_5080": 10752.0, "nvidia_rtx_5090": 20070.4}, "model_id": "ibm/granite-embedding-125m-english", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 54.0, "mteb_classification": 67.5, "mteb_clustering": 45.5, "mteb_overall": 61.5, "mteb_pair_classification": 83.0, "mteb_reranking": 55.0, "mteb_retrieval": 57.8, "mteb_sts": 80.5, "mteb_summarization": 28.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite embedding 278M multilingual", "fits": {"amd_rx_7900_xt": 4028.8, "amd_rx_7900_xtx": 4834.5, "amd_rx_9070_xt": 3248.2, "apple_m1_max": 2014.4, "apple_m2_max": 2014.4, "apple_m2_ultra": 4028.8, "apple_m3_max": 2014.4, "apple_m3_ultra": 4124.5, "apple_m4": 604.3, "apple_m4_max": 2749.6, "apple_m4_pro": 1374.8, "nvidia_rtx_3060_12gb": 1812.9, "nvidia_rtx_3090": 4713.7, "nvidia_rtx_4060_ti_16gb": 1450.4, "nvidia_rtx_4070_ti_super": 3384.2, "nvidia_rtx_4080_super": 3706.5, "nvidia_rtx_4090": 5076.3, "nvidia_rtx_5080": 4834.5, "nvidia_rtx_5090": 9024.5}, "model_id": "ibm/granite-embedding-278m-multilingual", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 43.0, "miracl": 32.0, "mteb_classification": 60.0, "mteb_clustering": 37.5, "mteb_overall": 53.0, "mteb_pair_classification": 78.5, "mteb_reranking": 50.0, "mteb_retrieval": 48.5, "mteb_sts": 75.0, "mteb_summarization": 25.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite embedding 30M english", "fits": {"amd_rx_7900_xt": 37333.3, "amd_rx_7900_xtx": 44800.0, "amd_rx_9070_xt": 30100.0, "apple_m1_max": 18666.7, "apple_m2_max": 18666.7, "apple_m2_ultra": 37333.3, "apple_m3_max": 18666.7, "apple_m3_ultra": 38220.0, "apple_m4": 5600.0, "apple_m4_max": 25480.0, "apple_m4_pro": 12740.0, "nvidia_rtx_3060_12gb": 16800.0, "nvidia_rtx_3090": 43680.0, "nvidia_rtx_4060_ti_16gb": 13440.0, "nvidia_rtx_4070_ti_super": 31360.0, "nvidia_rtx_4080_super": 34346.7, "nvidia_rtx_4090": 47040.0, "nvidia_rtx_5080": 44800.0, "nvidia_rtx_5090": 83626.7}, "model_id": "ibm/granite-embedding-30m-english", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 43.0, "miracl": 32.0, "mteb_classification": 60.0, "mteb_clustering": 37.5, "mteb_overall": 53.0, "mteb_pair_classification": 78.5, "mteb_reranking": 50.0, "mteb_retrieval": 48.5, "mteb_sts": 75.0, "mteb_summarization": 25.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite embedding 30M sparse", "fits": {"amd_rx_7900_xt": 37333.3, "amd_rx_7900_xtx": 44800.0, "amd_rx_9070_xt": 30100.0, "apple_m1_max": 18666.7, "apple_m2_max": 18666.7, "apple_m2_ultra": 37333.3, "apple_m3_max": 18666.7, "apple_m3_ultra": 38220.0, "apple_m4": 5600.0, "apple_m4_max": 25480.0, "apple_m4_pro": 12740.0, "nvidia_rtx_3060_12gb": 16800.0, "nvidia_rtx_3090": 43680.0, "nvidia_rtx_4060_ti_16gb": 13440.0, "nvidia_rtx_4070_ti_super": 31360.0, "nvidia_rtx_4080_super": 34346.7, "nvidia_rtx_4090": 47040.0, "nvidia_rtx_5080": 44800.0, "nvidia_rtx_5090": 83626.7}, "model_id": "ibm/granite-embedding-30m-sparse", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 53.5, "miracl": 46.0, "mteb_classification": 69.5, "mteb_clustering": 47.5, "mteb_overall": 63.5, "mteb_pair_classification": 84.5, "mteb_reranking": 56.5, "mteb_retrieval": 60.2, "mteb_sts": 82.0, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite embedding english r2", "fits": {"amd_rx_7900_xt": 7516.8, "amd_rx_7900_xtx": 9020.1, "amd_rx_9070_xt": 6060.4, "apple_m1_max": 3758.4, "apple_m2_max": 3758.4, "apple_m2_ultra": 7516.8, "apple_m3_max": 3758.4, "apple_m3_ultra": 7695.3, "apple_m4": 1127.5, "apple_m4_max": 5130.2, "apple_m4_pro": 2565.1, "nvidia_rtx_3060_12gb": 3382.6, "nvidia_rtx_3090": 8794.6, "nvidia_rtx_4060_ti_16gb": 2706.0, "nvidia_rtx_4070_ti_super": 6314.1, "nvidia_rtx_4080_super": 6915.4, "nvidia_rtx_4090": 9471.1, "nvidia_rtx_5080": 9020.1, "nvidia_rtx_5090": 16837.6}, "model_id": "ibm/granite-embedding-english-r2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 53.5, "miracl": 46.0, "mteb_classification": 69.5, "mteb_clustering": 47.5, "mteb_overall": 63.5, "mteb_pair_classification": 84.5, "mteb_reranking": 56.5, "mteb_retrieval": 60.2, "mteb_sts": 82.0, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite embedding small english r2", "fits": {"amd_rx_7900_xt": 23829.8, "amd_rx_7900_xtx": 28595.7, "amd_rx_9070_xt": 19212.8, "apple_m1_max": 11914.9, "apple_m2_max": 11914.9, "apple_m2_ultra": 23829.8, "apple_m3_max": 11914.9, "apple_m3_ultra": 24395.7, "apple_m4": 3574.5, "apple_m4_max": 16263.8, "apple_m4_pro": 8131.9, "nvidia_rtx_3060_12gb": 10723.4, "nvidia_rtx_3090": 27880.9, "nvidia_rtx_4060_ti_16gb": 8578.7, "nvidia_rtx_4070_ti_super": 20017.0, "nvidia_rtx_4080_super": 21923.4, "nvidia_rtx_4090": 30025.5, "nvidia_rtx_5080": 28595.7, "nvidia_rtx_5090": 53378.7}, "model_id": "ibm/granite-embedding-small-english-r2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "IBM", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite guardian 3.2 5B", "fits": {"amd_rx_7900_xt": 224.0, "amd_rx_7900_xtx": 268.8, "amd_rx_9070_xt": 180.6, "apple_m1_max": 112.0, "apple_m2_max": 112.0, "apple_m2_ultra": 224.0, "apple_m3_max": 112.0, "apple_m3_ultra": 229.3, "apple_m4": 33.6, "apple_m4_max": 152.9, "apple_m4_pro": 76.4, "nvidia_rtx_3060_12gb": 100.8, "nvidia_rtx_3090": 262.1, "nvidia_rtx_4060_ti_16gb": 80.6, "nvidia_rtx_4070_ti_super": 188.2, "nvidia_rtx_4080_super": 206.1, "nvidia_rtx_4090": 282.2, "nvidia_rtx_5080": 268.8, "nvidia_rtx_5090": 501.8}, "model_id": "ibm/granite-guardian-3-2-5b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite guardian 3.3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "ibm/granite-guardian-3-3-8b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite speech 3.3 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "ibm/granite-speech-3-3-2b", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "granite speech 3.3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "ibm/granite-speech-3-3-8b", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "granite timeseries flowstate r1", "fits": {}, "model_id": "ibm/granite-timeseries-flowstate-r1", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "granite timeseries patchtst", "fits": {}, "model_id": "ibm/granite-timeseries-patchtst", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "granite timeseries tspulse r1", "fits": {}, "model_id": "ibm/granite-timeseries-tspulse-r1", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "granite timeseries ttm r1", "fits": {}, "model_id": "ibm/granite-timeseries-ttm-r1", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "granite timeseries ttm r2", "fits": {}, "model_id": "ibm/granite-timeseries-ttm-r2", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "granite vision 3.3 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "ibm/granite-vision-3-3-2b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "granitelib rag r1.0", "fits": {}, "model_id": "ibm/granitelib-rag-r1-0", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "IBM", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.25, "display_name": "Mercury 2", "fits": {}, "model_id": "inception/mercury-2", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Inception (Mercury)", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.25, "display_name": "Mercury Coder", "fits": {}, "model_id": "inception/mercury-coder", "model_subtypes": [], "model_type": "llm-code", "open_weights": false, "provider": "Inception (Mercury)", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 128000, "cost_input": 0.25, "display_name": "Mercury Edit", "fits": {}, "model_id": "inception/mercury-edit", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Inception (Mercury)", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.25, "display_name": "Mercury", "fits": {}, "model_id": "inception/mercury", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Inception (Mercury)", "scores_as_of": null}, {"benchmark_scores": {"beir": 48.8, "miracl": 37.0, "mteb_classification": 65.8, "mteb_clustering": 43.8, "mteb_overall": 59.8, "mteb_pair_classification": 83.5, "mteb_reranking": 55.0, "mteb_retrieval": 55.5, "mteb_sts": 81.0, "mteb_summarization": 28.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "e5 base v2", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "intfloat/e5-base-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 47.5, "miracl": 35.0, "mteb_classification": 64.8, "mteb_clustering": 42.5, "mteb_overall": 58.5, "mteb_pair_classification": 83.0, "mteb_reranking": 54.0, "mteb_retrieval": 54.2, "mteb_sts": 80.5, "mteb_summarization": 28.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "e5 base", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "intfloat/e5-base", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 49.8, "miracl": 37.0, "mteb_classification": 66.2, "mteb_clustering": 44.2, "mteb_overall": 60.2, "mteb_pair_classification": 84.5, "mteb_reranking": 56.0, "mteb_retrieval": 56.5, "mteb_sts": 82.0, "mteb_summarization": 29.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "e5 large unsupervised", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "intfloat/e5-large-unsupervised", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 51.5, "miracl": 40.0, "mteb_classification": 68.2, "mteb_clustering": 46.5, "mteb_overall": 62.5, "mteb_pair_classification": 85.0, "mteb_reranking": 57.0, "mteb_retrieval": 58.8, "mteb_sts": 83.0, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "e5 large v2", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "intfloat/e5-large-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 50.5, "miracl": 38.5, "mteb_classification": 67.0, "mteb_clustering": 45.0, "mteb_overall": 61.0, "mteb_pair_classification": 84.5, "mteb_reranking": 56.0, "mteb_retrieval": 57.2, "mteb_sts": 82.0, "mteb_summarization": 29.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "e5 large", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "intfloat/e5-large", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 56.2, "miracl": 56.5, "mteb_classification": 71.5, "mteb_clustering": 50.8, "mteb_overall": 66.8, "mteb_pair_classification": 87.0, "mteb_reranking": 60.2, "mteb_retrieval": 63.2, "mteb_sts": 85.5, "mteb_summarization": 31.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "e5 mistral 7B instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "intfloat/e5-mistral-7b-instruct", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 46.5, "miracl": 34.5, "mteb_classification": 63.8, "mteb_clustering": 41.5, "mteb_overall": 57.5, "mteb_pair_classification": 82.0, "mteb_reranking": 53.0, "mteb_retrieval": 53.2, "mteb_sts": 79.0, "mteb_summarization": 27.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "e5 small v2", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "intfloat/e5-small-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 45.2, "miracl": 32.5, "mteb_classification": 62.5, "mteb_clustering": 40.0, "mteb_overall": 56.2, "mteb_pair_classification": 81.0, "mteb_reranking": 52.0, "mteb_retrieval": 51.8, "mteb_sts": 78.5, "mteb_summarization": 27.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "e5 small", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "intfloat/e5-small", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 48.5, "miracl": 52.5, "mteb_classification": 65.5, "mteb_clustering": 43.0, "mteb_overall": 59.0, "mteb_pair_classification": 83.0, "mteb_reranking": 54.2, "mteb_retrieval": 55.0, "mteb_sts": 80.0, "mteb_summarization": 28.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "multilingual e5 base", "fits": {"amd_rx_7900_xt": 4028.8, "amd_rx_7900_xtx": 4834.5, "amd_rx_9070_xt": 3248.2, "apple_m1_max": 2014.4, "apple_m2_max": 2014.4, "apple_m2_ultra": 4028.8, "apple_m3_max": 2014.4, "apple_m3_ultra": 4124.5, "apple_m4": 604.3, "apple_m4_max": 2749.6, "apple_m4_pro": 1374.8, "nvidia_rtx_3060_12gb": 1812.9, "nvidia_rtx_3090": 4713.7, "nvidia_rtx_4060_ti_16gb": 1450.4, "nvidia_rtx_4070_ti_super": 3384.2, "nvidia_rtx_4080_super": 3706.5, "nvidia_rtx_4090": 5076.3, "nvidia_rtx_5080": 4834.5, "nvidia_rtx_5090": 9024.5}, "model_id": "intfloat/multilingual-e5-base", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 53.5, "miracl": 58.2, "mteb_classification": 69.8, "mteb_clustering": 48.5, "mteb_overall": 64.5, "mteb_pair_classification": 86.2, "mteb_reranking": 58.6, "mteb_retrieval": 61.2, "mteb_sts": 84.8, "mteb_summarization": 30.4}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "multilingual e5 large instruct", "fits": {"amd_rx_7900_xt": 2000.0, "amd_rx_7900_xtx": 2400.0, "amd_rx_9070_xt": 1612.5, "apple_m1_max": 1000.0, "apple_m2_max": 1000.0, "apple_m2_ultra": 2000.0, "apple_m3_max": 1000.0, "apple_m3_ultra": 2047.5, "apple_m4": 300.0, "apple_m4_max": 1365.0, "apple_m4_pro": 682.5, "nvidia_rtx_3060_12gb": 900.0, "nvidia_rtx_3090": 2340.0, "nvidia_rtx_4060_ti_16gb": 720.0, "nvidia_rtx_4070_ti_super": 1680.0, "nvidia_rtx_4080_super": 1840.0, "nvidia_rtx_4090": 2520.0, "nvidia_rtx_5080": 2400.0, "nvidia_rtx_5090": 4480.0}, "model_id": "intfloat/multilingual-e5-large-instruct", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 50.8, "miracl": 55.1, "mteb_classification": 67.3, "mteb_clustering": 45.2, "mteb_overall": 61.2, "mteb_pair_classification": 84.8, "mteb_reranking": 56.5, "mteb_retrieval": 57.5, "mteb_sts": 82.5, "mteb_summarization": 29.2}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "multilingual e5 large", "fits": {"amd_rx_7900_xt": 2000.0, "amd_rx_7900_xtx": 2400.0, "amd_rx_9070_xt": 1612.5, "apple_m1_max": 1000.0, "apple_m2_max": 1000.0, "apple_m2_ultra": 2000.0, "apple_m3_max": 1000.0, "apple_m3_ultra": 2047.5, "apple_m4": 300.0, "apple_m4_max": 1365.0, "apple_m4_pro": 682.5, "nvidia_rtx_3060_12gb": 900.0, "nvidia_rtx_3090": 2340.0, "nvidia_rtx_4060_ti_16gb": 720.0, "nvidia_rtx_4070_ti_super": 1680.0, "nvidia_rtx_4080_super": 1840.0, "nvidia_rtx_4090": 2520.0, "nvidia_rtx_5080": 2400.0, "nvidia_rtx_5090": 4480.0}, "model_id": "intfloat/multilingual-e5-large", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 46.2, "miracl": 50.0, "mteb_classification": 63.5, "mteb_clustering": 41.0, "mteb_overall": 57.2, "mteb_pair_classification": 81.5, "mteb_reranking": 52.5, "mteb_retrieval": 53.0, "mteb_sts": 78.0, "mteb_summarization": 27.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "multilingual e5 small", "fits": {"amd_rx_7900_xt": 9491.5, "amd_rx_7900_xtx": 11389.8, "amd_rx_9070_xt": 7652.5, "apple_m1_max": 4745.8, "apple_m2_max": 4745.8, "apple_m2_ultra": 9491.5, "apple_m3_max": 4745.8, "apple_m3_ultra": 9716.9, "apple_m4": 1423.7, "apple_m4_max": 6478.0, "apple_m4_pro": 3239.0, "nvidia_rtx_3060_12gb": 4271.2, "nvidia_rtx_3090": 11105.1, "nvidia_rtx_4060_ti_16gb": 3416.9, "nvidia_rtx_4070_ti_super": 7972.9, "nvidia_rtx_4080_super": 8732.2, "nvidia_rtx_4090": 11959.3, "nvidia_rtx_5080": 11389.8, "nvidia_rtx_5090": 21261.0}, "model_id": "intfloat/multilingual-e5-small", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "intfloat", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "jina bert flash implementation", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "jina/jina-bert-flash-implementation", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Jina AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina clip v1", "fits": {"amd_rx_7900_xt": 5022.4, "amd_rx_7900_xtx": 6026.9, "amd_rx_9070_xt": 4049.3, "apple_m1_max": 2511.2, "apple_m2_max": 2511.2, "apple_m2_ultra": 5022.4, "apple_m3_max": 2511.2, "apple_m3_ultra": 5141.7, "apple_m4": 753.4, "apple_m4_max": 3427.8, "apple_m4_pro": 1713.9, "nvidia_rtx_3060_12gb": 2260.1, "nvidia_rtx_3090": 5876.2, "nvidia_rtx_4060_ti_16gb": 1808.1, "nvidia_rtx_4070_ti_super": 4218.8, "nvidia_rtx_4080_super": 4620.6, "nvidia_rtx_4090": 6328.3, "nvidia_rtx_5080": 6026.9, "nvidia_rtx_5090": 11250.2}, "model_id": "jina/jina-clip-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": null}, {"benchmark_scores": {"miracl": 61.2, "mteb_pair_classification": 84.0, "mteb_reranking": 58.1, "mteb_sts": 85.8, "mteb_summarization": 29.7}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina clip v2", "fits": {"amd_rx_7900_xt": 1294.8, "amd_rx_7900_xtx": 1553.8, "amd_rx_9070_xt": 1043.9, "apple_m1_max": 647.4, "apple_m2_max": 647.4, "apple_m2_ultra": 1294.8, "apple_m3_max": 647.4, "apple_m3_ultra": 1325.5, "apple_m4": 194.2, "apple_m4_max": 883.7, "apple_m4_pro": 441.8, "nvidia_rtx_3060_12gb": 582.7, "nvidia_rtx_3090": 1514.9, "nvidia_rtx_4060_ti_16gb": 466.1, "nvidia_rtx_4070_ti_super": 1087.6, "nvidia_rtx_4080_super": 1191.2, "nvidia_rtx_4090": 1631.4, "nvidia_rtx_5080": 1553.8, "nvidia_rtx_5090": 2900.3}, "model_id": "jina/jina-clip-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina code embeddings 0.5B", "fits": {"amd_rx_7900_xt": 2240.0, "amd_rx_7900_xtx": 2688.0, "amd_rx_9070_xt": 1806.0, "apple_m1_max": 1120.0, "apple_m2_max": 1120.0, "apple_m2_ultra": 2240.0, "apple_m3_max": 1120.0, "apple_m3_ultra": 2293.2, "apple_m4": 336.0, "apple_m4_max": 1528.8, "apple_m4_pro": 764.4, "nvidia_rtx_3060_12gb": 1008.0, "nvidia_rtx_3090": 2620.8, "nvidia_rtx_4060_ti_16gb": 806.4, "nvidia_rtx_4070_ti_super": 1881.6, "nvidia_rtx_4080_super": 2060.8, "nvidia_rtx_4090": 2822.4, "nvidia_rtx_5080": 2688.0, "nvidia_rtx_5090": 5017.6}, "model_id": "jina/jina-code-embeddings-0-5b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": null}, {"benchmark_scores": {"miracl": 61.2, "mteb_pair_classification": 84.0, "mteb_reranking": 58.1, "mteb_sts": 85.8, "mteb_summarization": 29.7}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina code embeddings 1.5B", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "jina/jina-code-embeddings-1-5b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "jina colbert v2", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "jina/jina-colbert-v2", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Jina AI", "scores_as_of": null}, {"benchmark_scores": {"mteb_classification": 65.0, "mteb_clustering": 43.5, "mteb_overall": 59.5, "mteb_retrieval": 55.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "jina embeddings v2 base code", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "jina/jina-embeddings-v2-base-code", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_classification": 65.8, "mteb_clustering": 44.2, "mteb_overall": 60.2, "mteb_retrieval": 56.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "jina embeddings v2 base de", "fits": {"amd_rx_7900_xt": 6956.5, "amd_rx_7900_xtx": 8347.8, "amd_rx_9070_xt": 5608.7, "apple_m1_max": 3478.3, "apple_m2_max": 3478.3, "apple_m2_ultra": 6956.5, "apple_m3_max": 3478.3, "apple_m3_ultra": 7121.7, "apple_m4": 1043.5, "apple_m4_max": 4747.8, "apple_m4_pro": 2373.9, "nvidia_rtx_3060_12gb": 3130.4, "nvidia_rtx_3090": 8139.1, "nvidia_rtx_4060_ti_16gb": 2504.3, "nvidia_rtx_4070_ti_super": 5843.5, "nvidia_rtx_4080_super": 6400.0, "nvidia_rtx_4090": 8765.2, "nvidia_rtx_5080": 8347.8, "nvidia_rtx_5090": 15582.6}, "model_id": "jina/jina-embeddings-v2-base-de", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 49.5, "mteb_classification": 66.2, "mteb_clustering": 44.8, "mteb_overall": 60.8, "mteb_retrieval": 56.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "jina embeddings v2 base en", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "jina/jina-embeddings-v2-base-en", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 47.0, "mteb_classification": 64.2, "mteb_clustering": 42.2, "mteb_overall": 58.5, "mteb_retrieval": 54.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.01, "display_name": "jina embeddings v2 small en", "fits": {"amd_rx_7900_xt": 33939.4, "amd_rx_7900_xtx": 40727.3, "amd_rx_9070_xt": 27363.6, "apple_m1_max": 16969.7, "apple_m2_max": 16969.7, "apple_m2_ultra": 33939.4, "apple_m3_max": 16969.7, "apple_m3_ultra": 34745.5, "apple_m4": 5090.9, "apple_m4_max": 23163.6, "apple_m4_pro": 11581.8, "nvidia_rtx_3060_12gb": 15272.7, "nvidia_rtx_3090": 39709.1, "nvidia_rtx_4060_ti_16gb": 12218.2, "nvidia_rtx_4070_ti_super": 28509.1, "nvidia_rtx_4080_super": 31224.2, "nvidia_rtx_4090": 42763.6, "nvidia_rtx_5080": 40727.3, "nvidia_rtx_5090": 76024.2}, "model_id": "jina/jina-embeddings-v2-small-en", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 55.8, "miracl": 61.2, "mteb_classification": 71.2, "mteb_clustering": 50.2, "mteb_overall": 66.5, "mteb_pair_classification": 84.0, "mteb_reranking": 58.1, "mteb_retrieval": 63.8, "mteb_sts": 85.8, "mteb_summarization": 29.7}, "capability_tiers": {}, "context_window": null, "cost_input": 0.02, "display_name": "jina embeddings v3", "fits": {"amd_rx_7900_xt": 1958.0, "amd_rx_7900_xtx": 2349.7, "amd_rx_9070_xt": 1578.7, "apple_m1_max": 979.0, "apple_m2_max": 979.0, "apple_m2_ultra": 1958.0, "apple_m3_max": 979.0, "apple_m3_ultra": 2004.5, "apple_m4": 293.7, "apple_m4_max": 1336.4, "apple_m4_pro": 668.2, "nvidia_rtx_3060_12gb": 881.1, "nvidia_rtx_3090": 2290.9, "nvidia_rtx_4060_ti_16gb": 704.9, "nvidia_rtx_4070_ti_super": 1644.8, "nvidia_rtx_4080_super": 1801.4, "nvidia_rtx_4090": 2467.1, "nvidia_rtx_5080": 2349.7, "nvidia_rtx_5090": 4386.0}, "model_id": "jina/jina-embeddings-v3", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 57.0, "miracl": 63.5, "mteb_classification": 72.0, "mteb_clustering": 51.5, "mteb_overall": 67.8, "mteb_pair_classification": 84.5, "mteb_reranking": 59.5, "mteb_retrieval": 65.0, "mteb_sts": 86.2, "mteb_summarization": 30.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.02, "display_name": "jina embeddings v4 vllm retrieval", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "jina/jina-embeddings-v4-vllm-retrieval", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 57.0, "miracl": 63.5, "mteb_classification": 72.0, "mteb_clustering": 51.5, "mteb_overall": 67.8, "mteb_pair_classification": 84.5, "mteb_reranking": 59.5, "mteb_retrieval": 65.0, "mteb_sts": 86.2, "mteb_summarization": 30.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.02, "display_name": "jina embeddings v4", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "jina/jina-embeddings-v4", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 63.5, "mteb_pair_classification": 84.5, "mteb_reranking": 59.5, "mteb_sts": 86.2, "mteb_summarization": 30.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.02, "display_name": "jina embeddings v5 text nano", "fits": {"amd_rx_7900_xt": 4686.2, "amd_rx_7900_xtx": 5623.4, "amd_rx_9070_xt": 3778.2, "apple_m1_max": 2343.1, "apple_m2_max": 2343.1, "apple_m2_ultra": 4686.2, "apple_m3_max": 2343.1, "apple_m3_ultra": 4797.5, "apple_m4": 702.9, "apple_m4_max": 3198.3, "apple_m4_pro": 1599.2, "nvidia_rtx_3060_12gb": 2108.8, "nvidia_rtx_3090": 5482.8, "nvidia_rtx_4060_ti_16gb": 1687.0, "nvidia_rtx_4070_ti_super": 3936.4, "nvidia_rtx_4080_super": 4311.3, "nvidia_rtx_4090": 5904.6, "nvidia_rtx_5080": 5623.4, "nvidia_rtx_5090": 10497.1}, "model_id": "jina/jina-embeddings-v5-text-nano", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 63.5, "mteb_pair_classification": 84.5, "mteb_reranking": 59.5, "mteb_sts": 86.2, "mteb_summarization": 30.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.02, "display_name": "jina embeddings v5 text small retrieval", "fits": {"amd_rx_7900_xt": 1654.4, "amd_rx_7900_xtx": 1985.2, "amd_rx_9070_xt": 1333.8, "apple_m1_max": 827.2, "apple_m2_max": 827.2, "apple_m2_ultra": 1654.4, "apple_m3_max": 827.2, "apple_m3_ultra": 1693.6, "apple_m4": 248.2, "apple_m4_max": 1129.1, "apple_m4_pro": 564.5, "nvidia_rtx_3060_12gb": 744.5, "nvidia_rtx_3090": 1935.6, "nvidia_rtx_4060_ti_16gb": 595.6, "nvidia_rtx_4070_ti_super": 1389.7, "nvidia_rtx_4080_super": 1522.0, "nvidia_rtx_4090": 2084.5, "nvidia_rtx_5080": 1985.2, "nvidia_rtx_5090": 3705.8}, "model_id": "jina/jina-embeddings-v5-text-small-retrieval", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 63.5, "mteb_pair_classification": 84.5, "mteb_reranking": 59.5, "mteb_sts": 86.2, "mteb_summarization": 30.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.02, "display_name": "jina embeddings v5 text small", "fits": {"amd_rx_7900_xt": 1654.4, "amd_rx_7900_xtx": 1985.2, "amd_rx_9070_xt": 1333.8, "apple_m1_max": 827.2, "apple_m2_max": 827.2, "apple_m2_ultra": 1654.4, "apple_m3_max": 827.2, "apple_m3_ultra": 1693.6, "apple_m4": 248.2, "apple_m4_max": 1129.1, "apple_m4_pro": 564.5, "nvidia_rtx_3060_12gb": 744.5, "nvidia_rtx_3090": 1935.6, "nvidia_rtx_4060_ti_16gb": 595.6, "nvidia_rtx_4070_ti_super": 1389.7, "nvidia_rtx_4080_super": 1522.0, "nvidia_rtx_4090": 2084.5, "nvidia_rtx_5080": 1985.2, "nvidia_rtx_5090": 3705.8}, "model_id": "jina/jina-embeddings-v5-text-small", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina reranker m0", "fits": {"amd_rx_7900_xt": 466.7, "amd_rx_7900_xtx": 560.0, "amd_rx_9070_xt": 376.2, "apple_m1_max": 233.3, "apple_m2_max": 233.3, "apple_m2_ultra": 466.7, "apple_m3_max": 233.3, "apple_m3_ultra": 477.8, "apple_m4": 70.0, "apple_m4_max": 318.5, "apple_m4_pro": 159.2, "nvidia_rtx_3060_12gb": 210.0, "nvidia_rtx_3090": 546.0, "nvidia_rtx_4060_ti_16gb": 168.0, "nvidia_rtx_4070_ti_super": 392.0, "nvidia_rtx_4080_super": 429.3, "nvidia_rtx_4090": 588.0, "nvidia_rtx_5080": 560.0, "nvidia_rtx_5090": 1045.3}, "model_id": "jina/jina-reranker-m0", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "Jina AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina reranker v1 turbo en", "fits": {"amd_rx_7900_xt": 29473.7, "amd_rx_7900_xtx": 35368.4, "amd_rx_9070_xt": 23763.2, "apple_m1_max": 14736.8, "apple_m2_max": 14736.8, "apple_m2_ultra": 29473.7, "apple_m3_max": 14736.8, "apple_m3_ultra": 30173.7, "apple_m4": 4421.1, "apple_m4_max": 20115.8, "apple_m4_pro": 10057.9, "nvidia_rtx_3060_12gb": 13263.2, "nvidia_rtx_3090": 34484.2, "nvidia_rtx_4060_ti_16gb": 10610.5, "nvidia_rtx_4070_ti_super": 24757.9, "nvidia_rtx_4080_super": 27115.8, "nvidia_rtx_4090": 37136.8, "nvidia_rtx_5080": 35368.4, "nvidia_rtx_5090": 66021.1}, "model_id": "jina/jina-reranker-v1-turbo-en", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "Jina AI", "scores_as_of": null}, {"benchmark_scores": {"mteb_overall": 68.5, "mteb_retrieval": 66.2}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina reranker v2 base multilingual", "fits": {"amd_rx_7900_xt": 4028.8, "amd_rx_7900_xtx": 4834.5, "amd_rx_9070_xt": 3248.2, "apple_m1_max": 2014.4, "apple_m2_max": 2014.4, "apple_m2_ultra": 4028.8, "apple_m3_max": 2014.4, "apple_m3_ultra": 4124.5, "apple_m4": 604.3, "apple_m4_max": 2749.6, "apple_m4_pro": 1374.8, "nvidia_rtx_3060_12gb": 1812.9, "nvidia_rtx_3090": 4713.7, "nvidia_rtx_4060_ti_16gb": 1450.4, "nvidia_rtx_4070_ti_super": 3384.2, "nvidia_rtx_4080_super": 3706.5, "nvidia_rtx_4090": 5076.3, "nvidia_rtx_5080": 4834.5, "nvidia_rtx_5090": 9024.5}, "model_id": "jina/jina-reranker-v2-base-multilingual", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "Jina AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "jina reranker v3", "fits": {"amd_rx_7900_xt": 1958.0, "amd_rx_7900_xtx": 2349.7, "amd_rx_9070_xt": 1578.7, "apple_m1_max": 979.0, "apple_m2_max": 979.0, "apple_m2_ultra": 1958.0, "apple_m3_max": 979.0, "apple_m3_ultra": 2004.5, "apple_m4": 293.7, "apple_m4_max": 1336.4, "apple_m4_pro": 668.2, "nvidia_rtx_3060_12gb": 881.1, "nvidia_rtx_3090": 2290.9, "nvidia_rtx_4060_ti_16gb": 704.9, "nvidia_rtx_4070_ti_super": 1644.8, "nvidia_rtx_4080_super": 1801.4, "nvidia_rtx_4090": 2467.1, "nvidia_rtx_5080": 2349.7, "nvidia_rtx_5090": 4386.0}, "model_id": "jina/jina-reranker-v3", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "Jina AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "align base", "fits": {"amd_rx_7900_xt": 2333.3, "amd_rx_7900_xtx": 2800.0, "amd_rx_9070_xt": 1881.2, "apple_m1_max": 1166.7, "apple_m2_max": 1166.7, "apple_m2_ultra": 2333.3, "apple_m3_max": 1166.7, "apple_m3_ultra": 2388.8, "apple_m4": 350.0, "apple_m4_max": 1592.5, "apple_m4_pro": 796.2, "nvidia_rtx_3060_12gb": 1050.0, "nvidia_rtx_3090": 2730.0, "nvidia_rtx_4060_ti_16gb": 840.0, "nvidia_rtx_4070_ti_super": 1960.0, "nvidia_rtx_4080_super": 2146.7, "nvidia_rtx_4090": 2940.0, "nvidia_rtx_5080": 2800.0, "nvidia_rtx_5090": 5226.7}, "model_id": "kakao/align-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Kakao Brain", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "LFM2 24B A2B GGUF", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "liquid/lfm2-24b-a2b-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "LFM2.5 1.2B Instruct GGUF", "fits": {"amd_rx_7900_xt": 933.3, "amd_rx_7900_xtx": 1120.0, "amd_rx_9070_xt": 752.5, "apple_m1_max": 466.7, "apple_m2_max": 466.7, "apple_m2_ultra": 933.3, "apple_m3_max": 466.7, "apple_m3_ultra": 955.5, "apple_m4": 140.0, "apple_m4_max": 637.0, "apple_m4_pro": 318.5, "nvidia_rtx_3060_12gb": 420.0, "nvidia_rtx_3090": 1092.0, "nvidia_rtx_4060_ti_16gb": 336.0, "nvidia_rtx_4070_ti_super": 784.0, "nvidia_rtx_4080_super": 858.7, "nvidia_rtx_4090": 1176.0, "nvidia_rtx_5080": 1120.0, "nvidia_rtx_5090": 2090.7}, "model_id": "liquid/lfm2-5-1-2b-instruct-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "LFM2.5 1.2B Instruct", "fits": {"amd_rx_7900_xt": 933.3, "amd_rx_7900_xtx": 1120.0, "amd_rx_9070_xt": 752.5, "apple_m1_max": 466.7, "apple_m2_max": 466.7, "apple_m2_ultra": 933.3, "apple_m3_max": 466.7, "apple_m3_ultra": 955.5, "apple_m4": 140.0, "apple_m4_max": 637.0, "apple_m4_pro": 318.5, "nvidia_rtx_3060_12gb": 420.0, "nvidia_rtx_3090": 1092.0, "nvidia_rtx_4060_ti_16gb": 336.0, "nvidia_rtx_4070_ti_super": 784.0, "nvidia_rtx_4080_super": 858.7, "nvidia_rtx_4090": 1176.0, "nvidia_rtx_5080": 1120.0, "nvidia_rtx_5090": 2090.7}, "model_id": "liquid/lfm2-5-1-2b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "LFM2.5 1.2B Thinking GGUF", "fits": {"amd_rx_7900_xt": 933.3, "amd_rx_7900_xtx": 1120.0, "amd_rx_9070_xt": 752.5, "apple_m1_max": 466.7, "apple_m2_max": 466.7, "apple_m2_ultra": 933.3, "apple_m3_max": 466.7, "apple_m3_ultra": 955.5, "apple_m4": 140.0, "apple_m4_max": 637.0, "apple_m4_pro": 318.5, "nvidia_rtx_3060_12gb": 420.0, "nvidia_rtx_3090": 1092.0, "nvidia_rtx_4060_ti_16gb": 336.0, "nvidia_rtx_4070_ti_super": 784.0, "nvidia_rtx_4080_super": 858.7, "nvidia_rtx_4090": 1176.0, "nvidia_rtx_5080": 1120.0, "nvidia_rtx_5090": 2090.7}, "model_id": "liquid/lfm2-5-1-2b-thinking-gguf", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "LFM2.5 Audio 1.5B GGUF LEAP", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "liquid/lfm2-5-audio-1-5b-gguf-leap", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "LFM2.5 VL 1.6B GGUF", "fits": {"amd_rx_7900_xt": 700.0, "amd_rx_7900_xtx": 840.0, "amd_rx_9070_xt": 564.4, "apple_m1_max": 350.0, "apple_m2_max": 350.0, "apple_m2_ultra": 700.0, "apple_m3_max": 350.0, "apple_m3_ultra": 716.6, "apple_m4": 105.0, "apple_m4_max": 477.7, "apple_m4_pro": 238.9, "nvidia_rtx_3060_12gb": 315.0, "nvidia_rtx_3090": 819.0, "nvidia_rtx_4060_ti_16gb": 252.0, "nvidia_rtx_4070_ti_super": 588.0, "nvidia_rtx_4080_super": 644.0, "nvidia_rtx_4090": 882.0, "nvidia_rtx_5080": 840.0, "nvidia_rtx_5090": 1568.0}, "model_id": "liquid/lfm2-5-vl-1-6b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "LFM2.5 VL 1.6B", "fits": {"amd_rx_7900_xt": 700.0, "amd_rx_7900_xtx": 840.0, "amd_rx_9070_xt": 564.4, "apple_m1_max": 350.0, "apple_m2_max": 350.0, "apple_m2_ultra": 700.0, "apple_m3_max": 350.0, "apple_m3_ultra": 716.6, "apple_m4": 105.0, "apple_m4_max": 477.7, "apple_m4_pro": 238.9, "nvidia_rtx_3060_12gb": 315.0, "nvidia_rtx_3090": 819.0, "nvidia_rtx_4060_ti_16gb": 252.0, "nvidia_rtx_4070_ti_super": 588.0, "nvidia_rtx_4080_super": 644.0, "nvidia_rtx_4090": 882.0, "nvidia_rtx_5080": 840.0, "nvidia_rtx_5090": 1568.0}, "model_id": "liquid/lfm2-5-vl-1-6b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "LFM2 8B A1B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "liquid/lfm2-8b-a1b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "LFM2 VL 3B GGUF", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "liquid/lfm2-vl-3b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "LFM2 VL 450M GGUF", "fits": {"amd_rx_7900_xt": 2488.9, "amd_rx_7900_xtx": 2986.7, "amd_rx_9070_xt": 2006.7, "apple_m1_max": 1244.4, "apple_m2_max": 1244.4, "apple_m2_ultra": 2488.9, "apple_m3_max": 1244.4, "apple_m3_ultra": 2548.0, "apple_m4": 373.3, "apple_m4_max": 1698.7, "apple_m4_pro": 849.3, "nvidia_rtx_3060_12gb": 1120.0, "nvidia_rtx_3090": 2912.0, "nvidia_rtx_4060_ti_16gb": 896.0, "nvidia_rtx_4070_ti_super": 2090.7, "nvidia_rtx_4080_super": 2289.8, "nvidia_rtx_4090": 3136.0, "nvidia_rtx_5080": 2986.7, "nvidia_rtx_5090": 5575.1}, "model_id": "liquid/lfm2-vl-450m-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Liquid AI", "scores_as_of": null}, {"benchmark_scores": {"mgsm": 92.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0, "display_name": "Cerebras-Llama-4-Maverick-17B-128E-Instruct", "fits": {"apple_m3_ultra": 2.9}, "model_id": "meta/cerebras-llama-4-maverick-17b-128e-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0, "display_name": "Cerebras-Llama-4-Scout-17B-16E-Instruct", "fits": {"apple_m2_max": 5.1, "apple_m2_ultra": 10.3, "apple_m3_max": 5.1, "apple_m3_ultra": 10.5, "apple_m4_max": 7.0}, "model_id": "meta/cerebras-llama-4-scout-17b-16e-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {"gsm8k": 21.6, "multipl_e": 58.8, "multipl_e_cpp": 58.2, "multipl_e_csharp": 55.2, "multipl_e_go": 55.2, "multipl_e_java": 64.5, "multipl_e_javascript": 62.1, "multipl_e_julia": 32.8, "multipl_e_kotlin": 45.2, "multipl_e_lua": 30.5, "multipl_e_perl": 25.2, "multipl_e_php": 52.8, "multipl_e_python": 68.5, "multipl_e_r": 28.5, "multipl_e_ruby": 38.5, "multipl_e_rust": 42.1, "multipl_e_scala": 32.5, "multipl_e_swift": 38.2, "multipl_e_typescript": 60.8, "winogrande": 73.4}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": null, "display_name": "CodeLlama 34B Instruct hf", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "meta/codellama-34b-instruct-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 5.5, "winogrande": 64.9}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": null, "display_name": "CodeLlama 7B hf", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "meta/codellama-7b-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"gsm8k": 5.5, "winogrande": 64.9}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": null, "display_name": "CodeLlama 7B Instruct hf", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "meta/codellama-7b-instruct-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0, "display_name": "Groq-Llama-4-Maverick-17B-128E-Instruct", "fits": {"apple_m3_ultra": 2.9}, "model_id": "meta/groq-llama-4-maverick-17b-128e-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {"bbh": 33.4, "gpqa_diamond": 23.2, "gsm8k": 15.2, "ifeval": 39.8, "math_500": 1.4, "mmlu_pro": 19.2, "musr": 40.1, "winogrande": 74.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 13B chat hf", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "meta/llama-2-13b-chat-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 41.3, "gpqa_diamond": 28.1, "gsm8k": 22.8, "ifeval": 24.8, "math_500": 1.5, "mmlu_pro": 23.8, "musr": 35.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 13B hf", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "meta/llama-2-13b-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 30.4, "gpqa_diamond": 26.4, "gsm8k": 26.7, "ifeval": 49.6, "math_500": 2.9, "mmlu_pro": 24.3, "musr": 36.9, "winogrande": 80.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 70B chat hf", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "meta/llama-2-70b-chat-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 54.7, "gpqa_diamond": 30.3, "gsm8k": 54.1, "ifeval": 24.1, "math_500": 3.2, "mmlu_pro": 37.2, "musr": 41.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 70B hf", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "meta/llama-2-70b-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 31.1, "gpqa_diamond": 25.3, "gsm8k": 7.4, "ifeval": 39.9, "math_500": 2.0, "mmlu_pro": 16.9, "musr": 36.8, "winogrande": 71.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 7B chat hf", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "meta/llama-2-7b-chat-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 35.0, "gpqa_diamond": 26.7, "gsm8k": 14.5, "ifeval": 25.2, "math_500": 1.7, "mmlu_pro": 18.6, "musr": 37.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 7B hf", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "meta/llama-2-7b-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 45.0, "gpqa_diamond": 49.8, "humaneval": 81.2, "ifeval": 80.0, "live_code_bench": 30.5, "math_500": 73.8, "mmlu_pro": 67.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 3.5, "display_name": "Llama 3.1 405B FP8", "fits": {"apple_m3_ultra": 2.8}, "model_id": "meta/llama-3-1-405b-fp8", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_hard_prompts": 1334.7, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "arena_elo_style_control": 1332.4, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 45.0, "gpqa_diamond": 49.8, "gsm8k": 96.8, "humaneval": 81.2, "ifeval": 80.0, "live_code_bench": 30.5, "math_500": 73.8, "mmlu_pro": 67.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 3.5, "display_name": "Llama 3.1 405B Instruct FP8", "fits": {"apple_m3_ultra": 2.8}, "model_id": "meta/llama-3-1-405b-instruct-fp8", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 39.3, "arena_elo_coding": 1260.0, "arena_elo_hard_prompts": 1340.1, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "arena_elo_style_control": 1334.2, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 45.0, "gpqa_diamond": 49.8, "gsm8k": 96.8, "humaneval": 81.2, "ifeval": 80.0, "live_code_bench": 30.5, "math_500": 73.8, "medqa": 82.9, "mmlu_pro": 67.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 3.5, "display_name": "Llama 3.1 405B Instruct", "fits": {"apple_m3_ultra": 2.8}, "model_id": "meta/llama-3-1-405b-instruct", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 80.0, "artificial_analysis_speed_index": 45.0, "gpqa_diamond": 49.8, "gsm8k": 96.8, "humaneval": 81.2, "ifeval": 80.0, "live_code_bench": 30.5, "math_500": 73.8, "mmlu_pro": 67.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 3.5, "display_name": "Llama 3.1 405B", "fits": {"apple_m3_ultra": 2.8}, "model_id": "meta/llama-3-1-405b", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 38.1, "arena_elo_coding": 1230.0, "arena_elo_hard_prompts": 1297.6, "arena_elo_math": 1220.0, "arena_elo_overall": 1240.0, "arena_elo_style_control": 1292.8, "artificial_analysis_quality_index": 75.0, "artificial_analysis_speed_index": 72.0, "bbh": 55.8, "bbq": 78.2, "finbench": 55.2, "gpqa_diamond": 42.5, "helm_safety": 82.5, "humaneval": 76.2, "ifeval": 76.0, "legalbench": 58.5, "live_code_bench": 23.2, "math_500": 64.5, "medqa": 65.2, "mmlu_astronomy": 70.2, "mmlu_biology": 79.5, "mmlu_business_ethics": 71.5, "mmlu_chemistry": 71.2, "mmlu_clinical_knowledge": 77.2, "mmlu_computer_science": 79.8, "mmlu_jurisprudence": 68.8, "mmlu_physics": 75.5, "mmlu_pro": 60.8, "mmlu_professional_accounting": 62.5, "mmlu_professional_law": 67.5, "multipl_e": 70.8, "multipl_e_cpp": 70.5, "multipl_e_csharp": 70.2, "multipl_e_go": 66.8, "multipl_e_java": 75.8, "multipl_e_javascript": 73.5, "multipl_e_julia": 50.5, "multipl_e_kotlin": 60.5, "multipl_e_lua": 43.8, "multipl_e_perl": 38.5, "multipl_e_php": 68.5, "multipl_e_python": 78.5, "multipl_e_r": 46.5, "multipl_e_ruby": 53.5, "multipl_e_rust": 58.2, "multipl_e_scala": 48.2, "multipl_e_swift": 53.2, "multipl_e_typescript": 72.5, "musr": 22.7, "toxigen": 86.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.88, "display_name": "Llama 3.1 70B Instruct", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "meta/llama-3-1-70b-instruct", "model_subtypes": ["llm-code"], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1230.0, "arena_elo_math": 1220.0, "arena_elo_overall": 1240.0, "artificial_analysis_quality_index": 75.0, "artificial_analysis_speed_index": 72.0, "bbh": 55.8, "bbq": 78.2, "finbench": 55.2, "gpqa_diamond": 42.5, "helm_safety": 82.5, "humaneval": 76.2, "ifeval": 76.0, "legalbench": 58.5, "live_code_bench": 23.2, "math_500": 64.5, "medqa": 65.2, "mmlu_astronomy": 70.2, "mmlu_biology": 79.5, "mmlu_business_ethics": 71.5, "mmlu_chemistry": 71.2, "mmlu_clinical_knowledge": 77.2, "mmlu_computer_science": 79.8, "mmlu_jurisprudence": 68.8, "mmlu_physics": 75.5, "mmlu_pro": 60.8, "mmlu_professional_accounting": 62.5, "mmlu_professional_law": 67.5, "multipl_e": 70.8, "multipl_e_cpp": 70.5, "multipl_e_csharp": 70.2, "multipl_e_go": 66.8, "multipl_e_java": 75.8, "multipl_e_javascript": 73.5, "multipl_e_julia": 50.5, "multipl_e_kotlin": 60.5, "multipl_e_lua": 43.8, "multipl_e_perl": 38.5, "multipl_e_php": 68.5, "multipl_e_python": 78.5, "multipl_e_r": 46.5, "multipl_e_ruby": 53.5, "multipl_e_rust": 58.2, "multipl_e_scala": 48.2, "multipl_e_swift": 53.2, "multipl_e_typescript": 72.5, "musr": 22.7, "toxigen": 86.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.88, "display_name": "Llama 3.1 70B", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "meta/llama-3-1-70b", "model_subtypes": ["llm-code"], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 22.9, "arena_elo_coding": 1160.0, "arena_elo_hard_prompts": 1221.8, "arena_elo_math": 1150.0, "arena_elo_overall": 1170.0, "arena_elo_style_control": 1211.0, "artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0, "bbh": 39.2, "bbq": 72.8, "gpqa_diamond": 28.5, "helm_safety": 78.5, "humaneval": 55.5, "ifeval": 62.0, "math_500": 42.0, "mmlu_pro": 40.5, "musr": 12.8, "toxigen": 82.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Llama 3.1 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/llama-3-1-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1160.0, "arena_elo_math": 1150.0, "arena_elo_overall": 1170.0, "artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0, "bbh": 39.2, "bbq": 72.8, "gpqa_diamond": 28.5, "helm_safety": 78.5, "humaneval": 55.5, "ifeval": 62.0, "math_500": 42.0, "mmlu_pro": 40.5, "musr": 12.8, "toxigen": 82.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Llama 3.1 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/llama-3-1-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbq": 78.2, "docvqa": 88.4, "helm_safety": 82.5, "mathvista": 51.5, "mmmu": 50.7, "toxigen": 85.8}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.18, "display_name": "Llama 3.2 11B Vision Instruct", "fits": {"amd_rx_7900_xt": 101.8, "amd_rx_7900_xtx": 122.2, "amd_rx_9070_xt": 82.1, "apple_m1_max": 50.9, "apple_m2_max": 50.9, "apple_m2_ultra": 101.8, "apple_m3_max": 50.9, "apple_m3_ultra": 104.2, "apple_m4": 15.3, "apple_m4_max": 69.5, "apple_m4_pro": 34.7, "nvidia_rtx_3060_12gb": 45.8, "nvidia_rtx_3090": 119.1, "nvidia_rtx_4060_ti_16gb": 36.7, "nvidia_rtx_4070_ti_super": 85.5, "nvidia_rtx_4080_super": 93.7, "nvidia_rtx_4090": 128.3, "nvidia_rtx_5080": 122.2, "nvidia_rtx_5090": 228.1}, "model_id": "meta/llama-3-2-11b-vision-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbq": 78.2, "docvqa": 88.4, "helm_safety": 82.5, "mathvista": 51.5, "mgsm": 68.9, "mmmu": 50.7, "toxigen": 85.8}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.18, "display_name": "Llama 3.2 11B Vision", "fits": {"amd_rx_7900_xt": 101.8, "amd_rx_7900_xtx": 122.2, "amd_rx_9070_xt": 82.1, "apple_m1_max": 50.9, "apple_m2_max": 50.9, "apple_m2_ultra": 101.8, "apple_m3_max": 50.9, "apple_m3_ultra": 104.2, "apple_m4": 15.3, "apple_m4_max": 69.5, "apple_m4_pro": 34.7, "nvidia_rtx_3060_12gb": 45.8, "nvidia_rtx_3090": 119.1, "nvidia_rtx_4060_ti_16gb": 36.7, "nvidia_rtx_4070_ti_super": 85.5, "nvidia_rtx_4080_super": 93.7, "nvidia_rtx_4090": 128.3, "nvidia_rtx_5080": 122.2, "nvidia_rtx_5090": 228.1}, "model_id": "meta/llama-3-2-11b-vision", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 34.8, "gpqa_diamond": 26.8, "ifeval": 58.1, "math_500": 8.2, "mmlu_pro": 17.4, "musr": 32.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.04, "display_name": "Llama 3.2 1B Instruct", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "meta/llama-3-2-1b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 31.1, "gpqa_diamond": 22.8, "ifeval": 14.8, "math_500": 1.2, "mmlu_pro": 12.0, "musr": 34.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.04, "display_name": "Llama 3.2 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "meta/llama-3-2-1b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 46.1, "gpqa_diamond": 27.9, "gsm8k": 77.7, "ifeval": 73.9, "math_500": 17.7, "medqa": 52.6, "mgsm": 58.2, "mmlu_pro": 31.9, "musr": 35.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.06, "display_name": "Llama 3.2 3B Instruct", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "meta/llama-3-2-3b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 39.1, "gpqa_diamond": 26.8, "gsm8k": 77.7, "ifeval": 13.4, "math_500": 1.9, "mgsm": 58.2, "mmlu_pro": 24.9, "musr": 35.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.06, "display_name": "Llama 3.2 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "meta/llama-3-2-3b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 92.3, "bbq": 79.8, "chartqa": 85.5, "docvqa": 90.1, "helm_safety": 84.5, "mathvista": 57.3, "mgsm": 86.9, "mmmu": 60.3, "toxigen": 87.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": 1.2, "display_name": "Llama 3.2 90B Vision Instruct", "fits": {"apple_m1_max": 6.2, "apple_m2_max": 6.2, "apple_m2_ultra": 12.4, "apple_m3_max": 6.2, "apple_m3_ultra": 12.7, "apple_m4_max": 8.5, "apple_m4_pro": 4.2}, "model_id": "meta/llama-3-2-90b-vision-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 92.3, "bbq": 79.8, "chartqa": 85.5, "docvqa": 90.1, "helm_safety": 84.5, "mathvista": 57.3, "mmmu": 60.3, "toxigen": 87.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": 1.2, "display_name": "Llama 3.2 90B Vision", "fits": {"apple_m1_max": 6.2, "apple_m2_max": 6.2, "apple_m2_ultra": 12.4, "apple_m3_max": 6.2, "apple_m3_ultra": 12.7, "apple_m4_max": 8.5, "apple_m4_pro": 4.2}, "model_id": "meta/llama-3-2-90b-vision", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 45.2, "alpaca_eval": 40.5, "arena_elo_coding": 1250.0, "arena_elo_hard_prompts": 1320.1, "arena_elo_math": 1240.0, "arena_elo_overall": 1260.0, "arena_elo_style_control": 1318.0, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 73.0, "bbh": 58.1, "bbq": 80.1, "gpqa_diamond": 50.2, "gsm8k": 95.0, "helm_safety": 84.2, "humaneval": 80.5, "ifeval": 78.5, "live_code_bench": 28.8, "math_500": 73.2, "mgsm": 91.1, "mmlu_astronomy": 71.5, "mmlu_biology": 80.2, "mmlu_business_ethics": 72.8, "mmlu_chemistry": 72.5, "mmlu_clinical_knowledge": 78.5, "mmlu_computer_science": 80.5, "mmlu_jurisprudence": 69.5, "mmlu_physics": 76.8, "mmlu_pro": 66.5, "mmlu_professional_accounting": 63.8, "mmlu_professional_law": 68.2, "mt_bench": 8.6, "multipl_e": 72.9, "multipl_e_cpp": 72.8, "multipl_e_csharp": 72.5, "multipl_e_go": 68.5, "multipl_e_java": 77.5, "multipl_e_javascript": 75.8, "multipl_e_julia": 52.2, "multipl_e_kotlin": 62.5, "multipl_e_lua": 45.8, "multipl_e_perl": 40.5, "multipl_e_php": 70.2, "multipl_e_python": 80.2, "multipl_e_r": 48.5, "multipl_e_ruby": 55.8, "multipl_e_rust": 60.5, "multipl_e_scala": 50.2, "multipl_e_swift": 55.5, "multipl_e_typescript": 74.8, "musr": 24.5, "swe_bench_verified": 25.8, "toxigen": 88.2, "wildbench": 68.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0, "display_name": "Llama-3.3-70B-Instruct", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "meta/llama-3-3-70b-instruct", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0, "display_name": "Llama-3.3-8B-Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/llama-3-3-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {"aider_polyglot": 15.6, "arena_elo_coding": 1320.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "bbq": 81.8, "chartqa": 90.0, "docvqa": 94.4, "gpqa_diamond": 62.5, "helm_safety": 86.5, "humaneval": 85.5, "ifeval": 83.0, "live_code_bench": 39.7, "math_500": 82.2, "mathvista": 73.7, "mgsm": 92.3, "mmlu_pro": 73.5, "mmmu": 73.4, "multipl_e_csharp": 75.5, "multipl_e_julia": 55.8, "multipl_e_kotlin": 66.8, "multipl_e_lua": 49.8, "multipl_e_perl": 44.5, "multipl_e_php": 73.2, "multipl_e_r": 52.2, "multipl_e_ruby": 60.2, "multipl_e_scala": 54.5, "multipl_e_swift": 59.8, "toxigen": 89.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0, "display_name": "Llama-4-Maverick-17B-128E-Instruct-FP8", "fits": {"apple_m3_ultra": 2.9}, "model_id": "meta/llama-4-maverick-17b-128e-instruct-fp8", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 15.6, "arena_elo_coding": 1320.0, "arena_elo_hard_prompts": 1338.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1310.0, "arena_elo_style_control": 1326.6, "arena_elo_vision": 1146.5, "bbq": 81.8, "chartqa": 90.0, "docvqa": 94.4, "gpqa_diamond": 62.5, "helm_safety": 86.5, "humaneval": 85.5, "ifeval": 83.0, "live_code_bench": 39.7, "math_500": 82.2, "mathvista": 73.7, "medqa": 78.4, "mgsm": 92.3, "mmlu_pro": 73.5, "mmmu": 73.4, "multipl_e_csharp": 75.5, "multipl_e_julia": 55.8, "multipl_e_kotlin": 66.8, "multipl_e_lua": 49.8, "multipl_e_perl": 44.5, "multipl_e_php": 73.2, "multipl_e_r": 52.2, "multipl_e_ruby": 60.2, "multipl_e_scala": 54.5, "multipl_e_swift": 59.8, "toxigen": 89.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2"}, "context_window": 1048576, "cost_input": 0.5, "display_name": "Llama 4 Maverick 17B 128E Instruct", "fits": {"amd_rx_7900_xt": 65.9, "amd_rx_7900_xtx": 79.1, "amd_rx_9070_xt": 53.1, "apple_m1_max": 32.9, "apple_m2_max": 32.9, "apple_m2_ultra": 65.9, "apple_m3_max": 32.9, "apple_m3_ultra": 67.4, "apple_m4": 9.9, "apple_m4_max": 45.0, "apple_m4_pro": 22.5, "nvidia_rtx_3060_12gb": 29.6, "nvidia_rtx_3090": 77.1, "nvidia_rtx_4060_ti_16gb": 23.7, "nvidia_rtx_4070_ti_super": 55.3, "nvidia_rtx_4080_super": 60.6, "nvidia_rtx_4090": 83.0, "nvidia_rtx_5080": 79.1, "nvidia_rtx_5090": 147.6}, "model_id": "meta/llama-4-maverick-17b-128e-instruct", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_math": 1270.0, "arena_elo_overall": 1280.0, "bbq": 80.5, "chartqa": 88.8, "docvqa": 94.4, "gpqa_diamond": 55.8, "helm_safety": 85.2, "humaneval": 80.2, "ifeval": 79.0, "live_code_bench": 29.9, "math_500": 76.5, "mathvista": 70.7, "mgsm": 90.6, "mmlu_pro": 68.2, "mmmu": 69.4, "multipl_e_csharp": 72.8, "multipl_e_julia": 52.5, "multipl_e_kotlin": 63.2, "multipl_e_lua": 46.5, "multipl_e_perl": 41.2, "multipl_e_php": 70.5, "multipl_e_r": 48.8, "multipl_e_ruby": 56.5, "multipl_e_scala": 50.8, "multipl_e_swift": 56.2, "toxigen": 88.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.0, "display_name": "Llama-4-Scout-17B-16E-Instruct-FP8", "fits": {"apple_m2_max": 5.1, "apple_m2_ultra": 10.3, "apple_m3_max": 5.1, "apple_m3_ultra": 10.5, "apple_m4_max": 7.0}, "model_id": "meta/llama-4-scout-17b-16e-instruct-fp8", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_hard_prompts": 1329.0, "arena_elo_math": 1270.0, "arena_elo_overall": 1280.0, "arena_elo_style_control": 1321.9, "arena_elo_vision": 1127.2, "bbq": 80.5, "chartqa": 88.8, "docvqa": 94.4, "gpqa_diamond": 55.8, "helm_safety": 85.2, "humaneval": 80.2, "ifeval": 79.0, "live_code_bench": 29.9, "math_500": 76.5, "mathvista": 70.7, "medqa": 52.0, "mgsm": 90.6, "mmlu_pro": 68.2, "mmmu": 69.4, "multipl_e_csharp": 72.8, "multipl_e_julia": 52.5, "multipl_e_kotlin": 63.2, "multipl_e_lua": 46.5, "multipl_e_perl": 41.2, "multipl_e_php": 70.5, "multipl_e_r": 48.8, "multipl_e_ruby": 56.5, "multipl_e_scala": 50.8, "multipl_e_swift": 56.2, "toxigen": 88.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2"}, "context_window": 1048576, "cost_input": 0.18, "display_name": "Llama 4 Scout 17B 16E Instruct", "fits": {"amd_rx_7900_xt": 65.9, "amd_rx_7900_xtx": 79.1, "amd_rx_9070_xt": 53.1, "apple_m1_max": 32.9, "apple_m2_max": 32.9, "apple_m2_ultra": 65.9, "apple_m3_max": 32.9, "apple_m3_ultra": 67.4, "apple_m4": 9.9, "apple_m4_max": 45.0, "apple_m4_pro": 22.5, "nvidia_rtx_3060_12gb": 29.6, "nvidia_rtx_3090": 77.1, "nvidia_rtx_4060_ti_16gb": 23.7, "nvidia_rtx_4070_ti_super": 55.3, "nvidia_rtx_4080_super": 60.6, "nvidia_rtx_4090": 83.0, "nvidia_rtx_5080": 79.1, "nvidia_rtx_5090": 147.6}, "model_id": "meta/llama-4-scout-17b-16e-instruct", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_math": 1270.0, "arena_elo_overall": 1280.0, "bbq": 80.5, "chartqa": 88.8, "docvqa": 94.4, "gpqa_diamond": 55.8, "helm_safety": 85.2, "humaneval": 80.2, "ifeval": 79.0, "live_code_bench": 29.9, "math_500": 76.5, "mathvista": 70.7, "mgsm": 90.6, "mmlu_pro": 68.2, "mmmu": 69.4, "multipl_e_csharp": 72.8, "multipl_e_julia": 52.5, "multipl_e_kotlin": 63.2, "multipl_e_lua": 46.5, "multipl_e_perl": 41.2, "multipl_e_php": 70.5, "multipl_e_r": 48.8, "multipl_e_ruby": 56.5, "multipl_e_scala": 50.8, "multipl_e_swift": 56.2, "toxigen": 88.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2"}, "context_window": 1048576, "cost_input": 0.18, "display_name": "Llama 4 Scout 17B 16E", "fits": {"amd_rx_7900_xt": 65.9, "amd_rx_7900_xtx": 79.1, "amd_rx_9070_xt": 53.1, "apple_m1_max": 32.9, "apple_m2_max": 32.9, "apple_m2_ultra": 65.9, "apple_m3_max": 32.9, "apple_m3_ultra": 67.4, "apple_m4": 9.9, "apple_m4_max": 45.0, "apple_m4_pro": 22.5, "nvidia_rtx_3060_12gb": 29.6, "nvidia_rtx_3090": 77.1, "nvidia_rtx_4060_ti_16gb": 23.7, "nvidia_rtx_4070_ti_super": 55.3, "nvidia_rtx_4080_super": 60.6, "nvidia_rtx_4090": 83.0, "nvidia_rtx_5080": 79.1, "nvidia_rtx_5090": 147.6}, "model_id": "meta/llama-4-scout-17b-16e", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Llama Guard 3 11B Vision", "fits": {"amd_rx_7900_xt": 101.8, "amd_rx_7900_xtx": 122.2, "amd_rx_9070_xt": 82.1, "apple_m1_max": 50.9, "apple_m2_max": 50.9, "apple_m2_ultra": 101.8, "apple_m3_max": 50.9, "apple_m3_ultra": 104.2, "apple_m4": 15.3, "apple_m4_max": 69.5, "apple_m4_pro": 34.7, "nvidia_rtx_3060_12gb": 45.8, "nvidia_rtx_3090": 119.1, "nvidia_rtx_4060_ti_16gb": 36.7, "nvidia_rtx_4070_ti_super": 85.5, "nvidia_rtx_4080_super": 93.7, "nvidia_rtx_4090": 128.3, "nvidia_rtx_5080": 122.2, "nvidia_rtx_5090": 228.1}, "model_id": "meta/llama-guard-3-11b-vision", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {"helm_safety": 92.8, "toxigen": 95.8}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Llama Guard 3 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "meta/llama-guard-3-1b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"helm_safety": 95.2, "toxigen": 97.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Llama Guard 3 8B INT8", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/llama-guard-3-8b-int8", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"helm_safety": 95.2, "toxigen": 97.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Llama Guard 3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/llama-guard-3-8b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Llama Guard 4 12B", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "meta/llama-guard-4-12b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Llama Prompt Guard 2 22M", "fits": {"amd_rx_7900_xt": 50909.1, "amd_rx_7900_xtx": 61090.9, "amd_rx_9070_xt": 41045.5, "apple_m1_max": 25454.5, "apple_m2_max": 25454.5, "apple_m2_ultra": 50909.1, "apple_m3_max": 25454.5, "apple_m3_ultra": 52118.2, "apple_m4": 7636.4, "apple_m4_max": 34745.5, "apple_m4_pro": 17372.7, "nvidia_rtx_3060_12gb": 22909.1, "nvidia_rtx_3090": 59563.6, "nvidia_rtx_4060_ti_16gb": 18327.3, "nvidia_rtx_4070_ti_super": 42763.6, "nvidia_rtx_4080_super": 46836.4, "nvidia_rtx_4090": 64145.5, "nvidia_rtx_5080": 61090.9, "nvidia_rtx_5090": 114036.4}, "model_id": "meta/llama-prompt-guard-2-22m", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Llama Prompt Guard 2 86M", "fits": {"amd_rx_7900_xt": 13023.3, "amd_rx_7900_xtx": 15627.9, "amd_rx_9070_xt": 10500.0, "apple_m1_max": 6511.6, "apple_m2_max": 6511.6, "apple_m2_ultra": 13023.3, "apple_m3_max": 6511.6, "apple_m3_ultra": 13332.6, "apple_m4": 1953.5, "apple_m4_max": 8888.4, "apple_m4_pro": 4444.2, "nvidia_rtx_3060_12gb": 5860.5, "nvidia_rtx_3090": 15237.2, "nvidia_rtx_4060_ti_16gb": 4688.4, "nvidia_rtx_4070_ti_super": 10939.5, "nvidia_rtx_4080_super": 11981.4, "nvidia_rtx_4090": 16409.3, "nvidia_rtx_5080": 15627.9, "nvidia_rtx_5090": 29172.1}, "model_id": "meta/llama-prompt-guard-2-86m", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "LlamaGuard 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "meta/llamaguard-7b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 71.4, "bbh": 65.5, "gpqa_diamond": 28.7, "gsm8k": 95.0, "hellaswag": 85.7, "ifeval": 81.0, "math_500": 24.5, "mgsm": 91.1, "mmlu_abstract_algebra": 51.0, "mmlu_anatomy": 77.0, "mmlu_astronomy": 93.4, "mmlu_business_ethics": 86.0, "mmlu_clinical_knowledge": 86.8, "mmlu_college_biology": 91.0, "mmlu_college_chemistry": 56.0, "mmlu_college_computer_science": 69.0, "mmlu_college_mathematics": 58.0, "mmlu_college_medicine": 81.5, "mmlu_college_physics": 59.8, "mmlu_computer_security": 82.0, "mmlu_conceptual_physics": 81.7, "mmlu_econometrics": 72.8, "mmlu_electrical_engineering": 77.9, "mmlu_elementary_mathematics": 70.9, "mmlu_formal_logic": 62.7, "mmlu_global_facts": 59.0, "mmlu_high_school_biology": 89.7, "mmlu_high_school_chemistry": 69.0, "mmlu_high_school_computer_science": 92.0, "mmlu_high_school_european_history": 86.7, "mmlu_high_school_geography": 92.9, "mmlu_high_school_government_and_politics": 98.4, "mmlu_high_school_macroeconomics": 83.8, "mmlu_high_school_mathematics": 54.4, "mmlu_high_school_microeconomics": 88.7, "mmlu_high_school_physics": 62.3, "mmlu_high_school_psychology": 94.1, "mmlu_high_school_statistics": 73.6, "mmlu_high_school_us_history": 92.2, "mmlu_high_school_world_history": 92.8, "mmlu_human_aging": 82.1, "mmlu_human_sexuality": 87.0, "mmlu_international_law": 92.6, "mmlu_jurisprudence": 88.0, "mmlu_logical_fallacies": 85.9, "mmlu_machine_learning": 70.5, "mmlu_management": 89.3, "mmlu_marketing": 93.6, "mmlu_medical_genetics": 92.0, "mmlu_miscellaneous": 91.8, "mmlu_moral_disputes": 84.4, "mmlu_moral_scenarios": 70.8, "mmlu_nutrition": 87.6, "mmlu_philosophy": 81.4, "mmlu_prehistory": 90.1, "mmlu_pro": 52.1, "mmlu_professional_accounting": 63.8, "mmlu_professional_law": 64.1, "mmlu_professional_medicine": 89.0, "mmlu_professional_psychology": 85.3, "mmlu_public_relations": 74.5, "mmlu_security_studies": 81.6, "mmlu_sociology": 92.0, "mmlu_us_foreign_policy": 92.0, "mmlu_virology": 56.6, "mmlu_world_religions": 90.1, "musr": 41.5, "truthfulqa": 61.8, "winogrande": 82.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Meta Llama 3 70B Instruct", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "meta/meta-llama-3-70b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 71.4, "bbh": 64.6, "gpqa_diamond": 39.8, "gsm8k": 85.4, "hellaswag": 85.7, "ifeval": 16.0, "math_500": 18.6, "mmlu_abstract_algebra": 51.0, "mmlu_anatomy": 77.0, "mmlu_astronomy": 93.4, "mmlu_business_ethics": 86.0, "mmlu_clinical_knowledge": 86.8, "mmlu_college_biology": 91.0, "mmlu_college_chemistry": 56.0, "mmlu_college_computer_science": 69.0, "mmlu_college_mathematics": 58.0, "mmlu_college_medicine": 81.5, "mmlu_college_physics": 59.8, "mmlu_computer_security": 82.0, "mmlu_conceptual_physics": 81.7, "mmlu_econometrics": 72.8, "mmlu_electrical_engineering": 77.9, "mmlu_elementary_mathematics": 70.9, "mmlu_formal_logic": 62.7, "mmlu_global_facts": 59.0, "mmlu_high_school_biology": 89.7, "mmlu_high_school_chemistry": 69.0, "mmlu_high_school_computer_science": 92.0, "mmlu_high_school_european_history": 86.7, "mmlu_high_school_geography": 92.9, "mmlu_high_school_government_and_politics": 98.4, "mmlu_high_school_macroeconomics": 83.8, "mmlu_high_school_mathematics": 54.4, "mmlu_high_school_microeconomics": 88.7, "mmlu_high_school_physics": 62.3, "mmlu_high_school_psychology": 94.1, "mmlu_high_school_statistics": 73.6, "mmlu_high_school_us_history": 92.2, "mmlu_high_school_world_history": 92.8, "mmlu_human_aging": 82.1, "mmlu_human_sexuality": 87.0, "mmlu_international_law": 92.6, "mmlu_jurisprudence": 88.0, "mmlu_logical_fallacies": 85.9, "mmlu_machine_learning": 70.5, "mmlu_management": 89.3, "mmlu_marketing": 93.6, "mmlu_medical_genetics": 92.0, "mmlu_miscellaneous": 91.8, "mmlu_moral_disputes": 84.4, "mmlu_moral_scenarios": 70.8, "mmlu_nutrition": 87.6, "mmlu_philosophy": 81.4, "mmlu_prehistory": 90.1, "mmlu_pro": 47.1, "mmlu_professional_accounting": 63.8, "mmlu_professional_law": 64.1, "mmlu_professional_medicine": 89.0, "mmlu_professional_psychology": 85.3, "mmlu_public_relations": 74.5, "mmlu_security_studies": 81.6, "mmlu_sociology": 92.0, "mmlu_us_foreign_policy": 92.0, "mmlu_virology": 56.6, "mmlu_world_religions": 90.1, "musr": 45.2, "truthfulqa": 61.8, "winogrande": 82.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Meta Llama 3 70B", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "meta/meta-llama-3-70b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 60.8, "bbh": 49.1, "gpqa_diamond": 29.3, "gsm8k": 68.7, "hellaswag": 78.6, "ifeval": 47.8, "math_500": 9.1, "mmlu_abstract_algebra": 32.0, "mmlu_anatomy": 65.2, "mmlu_astronomy": 70.4, "mmlu_business_ethics": 69.0, "mmlu_clinical_knowledge": 74.7, "mmlu_college_biology": 79.9, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 59.0, "mmlu_college_mathematics": 39.0, "mmlu_college_medicine": 63.6, "mmlu_college_physics": 50.0, "mmlu_computer_security": 77.0, "mmlu_conceptual_physics": 60.4, "mmlu_econometrics": 60.5, "mmlu_electrical_engineering": 62.8, "mmlu_elementary_mathematics": 44.4, "mmlu_formal_logic": 48.4, "mmlu_global_facts": 40.0, "mmlu_high_school_biology": 78.4, "mmlu_high_school_chemistry": 50.7, "mmlu_high_school_computer_science": 75.0, "mmlu_high_school_european_history": 74.5, "mmlu_high_school_geography": 84.3, "mmlu_high_school_government_and_politics": 91.2, "mmlu_high_school_macroeconomics": 65.9, "mmlu_high_school_mathematics": 40.0, "mmlu_high_school_microeconomics": 76.5, "mmlu_high_school_physics": 44.4, "mmlu_high_school_psychology": 83.7, "mmlu_high_school_statistics": 53.7, "mmlu_high_school_us_history": 85.8, "mmlu_high_school_world_history": 84.4, "mmlu_human_aging": 72.2, "mmlu_human_sexuality": 77.9, "mmlu_international_law": 81.8, "mmlu_jurisprudence": 77.8, "mmlu_logical_fallacies": 76.7, "mmlu_machine_learning": 54.5, "mmlu_management": 78.6, "mmlu_marketing": 90.6, "mmlu_medical_genetics": 80.0, "mmlu_miscellaneous": 79.8, "mmlu_moral_disputes": 74.9, "mmlu_moral_scenarios": 43.7, "mmlu_nutrition": 74.5, "mmlu_philosophy": 71.4, "mmlu_prehistory": 74.1, "mmlu_pro": 35.9, "mmlu_professional_accounting": 52.8, "mmlu_professional_law": 47.8, "mmlu_professional_medicine": 71.7, "mmlu_professional_psychology": 70.6, "mmlu_public_relations": 64.5, "mmlu_security_studies": 73.9, "mmlu_sociology": 86.6, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 51.2, "mmlu_world_religions": 77.8, "musr": 38.1, "truthfulqa": 51.6, "winogrande": 74.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Meta Llama 3 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/meta-llama-3-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 60.2, "bbh": 46.0, "gpqa_diamond": 30.5, "gsm8k": 45.2, "hellaswag": 82.2, "ifeval": 14.6, "math_500": 4.5, "mmlu_abstract_algebra": 35.0, "mmlu_anatomy": 69.6, "mmlu_astronomy": 67.8, "mmlu_business_ethics": 63.0, "mmlu_clinical_knowledge": 75.5, "mmlu_college_biology": 77.1, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 49.0, "mmlu_college_mathematics": 33.0, "mmlu_college_medicine": 63.6, "mmlu_college_physics": 49.0, "mmlu_computer_security": 80.0, "mmlu_conceptual_physics": 59.1, "mmlu_econometrics": 49.1, "mmlu_electrical_engineering": 64.8, "mmlu_elementary_mathematics": 43.4, "mmlu_formal_logic": 47.6, "mmlu_global_facts": 37.0, "mmlu_high_school_biology": 76.8, "mmlu_high_school_chemistry": 54.7, "mmlu_high_school_computer_science": 68.0, "mmlu_high_school_european_history": 77.0, "mmlu_high_school_geography": 80.3, "mmlu_high_school_government_and_politics": 89.1, "mmlu_high_school_macroeconomics": 65.1, "mmlu_high_school_mathematics": 40.0, "mmlu_high_school_microeconomics": 73.5, "mmlu_high_school_physics": 44.4, "mmlu_high_school_psychology": 84.8, "mmlu_high_school_statistics": 57.9, "mmlu_high_school_us_history": 85.8, "mmlu_high_school_world_history": 82.3, "mmlu_human_aging": 71.7, "mmlu_human_sexuality": 77.1, "mmlu_international_law": 85.1, "mmlu_jurisprudence": 75.0, "mmlu_logical_fallacies": 73.6, "mmlu_machine_learning": 53.6, "mmlu_management": 87.4, "mmlu_marketing": 89.7, "mmlu_medical_genetics": 79.0, "mmlu_miscellaneous": 83.0, "mmlu_moral_disputes": 74.0, "mmlu_moral_scenarios": 41.3, "mmlu_nutrition": 76.5, "mmlu_philosophy": 74.3, "mmlu_prehistory": 72.5, "mmlu_pro": 32.1, "mmlu_professional_accounting": 49.3, "mmlu_professional_law": 46.7, "mmlu_professional_medicine": 72.1, "mmlu_professional_psychology": 72.4, "mmlu_public_relations": 71.8, "mmlu_security_studies": 75.1, "mmlu_sociology": 85.6, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 56.0, "mmlu_world_religions": 83.0, "musr": 36.1, "truthfulqa": 42.9, "winogrande": 78.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Meta Llama 3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/meta-llama-3-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Meta", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Meta Llama Guard 2 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "meta/meta-llama-guard-2-8b", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {"arc_agi_2": 42.5, "artificial_analysis_quality_index": 52.0, "charxiv_reasoning": 86.4, "deepsearchqa": 74.8, "frontierscience_research": 38.3, "healthbench_hard": 42.8, "hle": 50.2, "ipho_2025_theory": 82.6, "medxpertqa_multimodal": 78.4, "terminal_bench_2": 59.0, "zerobench": 33.0}, "capability_tiers": {"coding": "tier-2", "coding:agentic_coding": "tier-2", "reasoning": "tier-1", "reasoning:causal": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:logical": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "reasoning:planning": "tier-1", "reasoning:scientific": "tier-1", "reasoning:self_correction": "tier-1", "reasoning:think_budget_control": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1", "tool_use:parallel_tool_calls": "tier-1"}, "context_window": 8192, "cost_input": null, "display_name": "Muse Spark", "fits": {}, "model_id": "meta/muse-spark", "model_subtypes": ["vlm"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Meta", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Prompt Guard 86M", "fits": {"amd_rx_7900_xt": 13023.3, "amd_rx_7900_xtx": 15627.9, "amd_rx_9070_xt": 10500.0, "apple_m1_max": 6511.6, "apple_m2_max": 6511.6, "apple_m2_ultra": 13023.3, "apple_m3_max": 6511.6, "apple_m3_ultra": 13332.6, "apple_m4": 1953.5, "apple_m4_max": 8888.4, "apple_m4_pro": 4444.2, "nvidia_rtx_3060_12gb": 5860.5, "nvidia_rtx_3090": 15237.2, "nvidia_rtx_4060_ti_16gb": 4688.4, "nvidia_rtx_4070_ti_super": 10939.5, "nvidia_rtx_4080_super": 11981.4, "nvidia_rtx_4090": 16409.3, "nvidia_rtx_5080": 15627.9, "nvidia_rtx_5090": 29172.1}, "model_id": "meta/prompt-guard-86m", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Meta", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "beit base patch16 224 pt22k ft22k", "fits": {"amd_rx_7900_xt": 12873.6, "amd_rx_7900_xtx": 15448.3, "amd_rx_9070_xt": 10379.3, "apple_m1_max": 6436.8, "apple_m2_max": 6436.8, "apple_m2_ultra": 12873.6, "apple_m3_max": 6436.8, "apple_m3_ultra": 13179.3, "apple_m4": 1931.0, "apple_m4_max": 8786.2, "apple_m4_pro": 4393.1, "nvidia_rtx_3060_12gb": 5793.1, "nvidia_rtx_3090": 15062.1, "nvidia_rtx_4060_ti_16gb": 4634.5, "nvidia_rtx_4070_ti_super": 10813.8, "nvidia_rtx_4080_super": 11843.7, "nvidia_rtx_4090": 16220.7, "nvidia_rtx_5080": 15448.3, "nvidia_rtx_5090": 28836.8}, "model_id": "microsoft/beit-base-patch16-224-pt22k-ft22k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "biogpt", "fits": {"amd_rx_7900_xt": 3227.7, "amd_rx_7900_xtx": 3873.2, "amd_rx_9070_xt": 2602.3, "apple_m1_max": 1613.8, "apple_m2_max": 1613.8, "apple_m2_ultra": 3227.7, "apple_m3_max": 1613.8, "apple_m3_ultra": 3304.3, "apple_m4": 484.1, "apple_m4_max": 2202.9, "apple_m4_pro": 1101.4, "nvidia_rtx_3060_12gb": 1452.4, "nvidia_rtx_3090": 3776.4, "nvidia_rtx_4060_ti_16gb": 1162.0, "nvidia_rtx_4070_ti_super": 2711.2, "nvidia_rtx_4080_super": 2969.5, "nvidia_rtx_4090": 4066.9, "nvidia_rtx_5080": 3873.2, "nvidia_rtx_5090": 7230.0}, "model_id": "microsoft/biogpt", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "BiomedCLIP PubMedBERT 256 vit base patch16 224", "fits": {"amd_rx_7900_xt": 5743.6, "amd_rx_7900_xtx": 6892.3, "amd_rx_9070_xt": 4630.8, "apple_m1_max": 2871.8, "apple_m2_max": 2871.8, "apple_m2_ultra": 5743.6, "apple_m3_max": 2871.8, "apple_m3_ultra": 5880.0, "apple_m4": 861.5, "apple_m4_max": 3920.0, "apple_m4_pro": 1960.0, "nvidia_rtx_3060_12gb": 2584.6, "nvidia_rtx_3090": 6720.0, "nvidia_rtx_4060_ti_16gb": 2067.7, "nvidia_rtx_4070_ti_super": 4824.6, "nvidia_rtx_4080_super": 5284.1, "nvidia_rtx_4090": 7236.9, "nvidia_rtx_5080": 6892.3, "nvidia_rtx_5090": 12865.6}, "model_id": "microsoft/biomedclip-pubmedbert-256-vit-base-patch16-224", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "BiomedNLP BiomedBERT base uncased abstract fulltext", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "microsoft/biomednlp-biomedbert-base-uncased-abstract-fulltext", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "BiomedNLP BiomedBERT base uncased abstract", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "microsoft/biomednlp-biomedbert-base-uncased-abstract", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "codebert base", "fits": {"amd_rx_7900_xt": 8960.0, "amd_rx_7900_xtx": 10752.0, "amd_rx_9070_xt": 7224.0, "apple_m1_max": 4480.0, "apple_m2_max": 4480.0, "apple_m2_ultra": 8960.0, "apple_m3_max": 4480.0, "apple_m3_ultra": 9172.8, "apple_m4": 1344.0, "apple_m4_max": 6115.2, "apple_m4_pro": 3057.6, "nvidia_rtx_3060_12gb": 4032.0, "nvidia_rtx_3090": 10483.2, "nvidia_rtx_4060_ti_16gb": 3225.6, "nvidia_rtx_4070_ti_super": 7526.4, "nvidia_rtx_4080_super": 8243.2, "nvidia_rtx_4090": 11289.6, "nvidia_rtx_5080": 10752.0, "nvidia_rtx_5090": 20070.4}, "model_id": "microsoft/codebert-base", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "deberta large mnli", "fits": {"amd_rx_7900_xt": 3200.0, "amd_rx_7900_xtx": 3840.0, "amd_rx_9070_xt": 2580.0, "apple_m1_max": 1600.0, "apple_m2_max": 1600.0, "apple_m2_ultra": 3200.0, "apple_m3_max": 1600.0, "apple_m3_ultra": 3276.0, "apple_m4": 480.0, "apple_m4_max": 2184.0, "apple_m4_pro": 1092.0, "nvidia_rtx_3060_12gb": 1440.0, "nvidia_rtx_3090": 3744.0, "nvidia_rtx_4060_ti_16gb": 1152.0, "nvidia_rtx_4070_ti_super": 2688.0, "nvidia_rtx_4080_super": 2944.0, "nvidia_rtx_4090": 4032.0, "nvidia_rtx_5080": 3840.0, "nvidia_rtx_5090": 7168.0}, "model_id": "microsoft/deberta-large-mnli", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "deberta v2 xlarge", "fits": {"amd_rx_7900_xt": 1244.4, "amd_rx_7900_xtx": 1493.3, "amd_rx_9070_xt": 1003.3, "apple_m1_max": 622.2, "apple_m2_max": 622.2, "apple_m2_ultra": 1244.4, "apple_m3_max": 622.2, "apple_m3_ultra": 1274.0, "apple_m4": 186.7, "apple_m4_max": 849.3, "apple_m4_pro": 424.7, "nvidia_rtx_3060_12gb": 560.0, "nvidia_rtx_3090": 1456.0, "nvidia_rtx_4060_ti_16gb": 448.0, "nvidia_rtx_4070_ti_super": 1045.3, "nvidia_rtx_4080_super": 1144.9, "nvidia_rtx_4090": 1568.0, "nvidia_rtx_5080": 1493.3, "nvidia_rtx_5090": 2787.6}, "model_id": "microsoft/deberta-v2-xlarge", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "deberta v3 base", "fits": {"amd_rx_7900_xt": 13023.3, "amd_rx_7900_xtx": 15627.9, "amd_rx_9070_xt": 10500.0, "apple_m1_max": 6511.6, "apple_m2_max": 6511.6, "apple_m2_ultra": 13023.3, "apple_m3_max": 6511.6, "apple_m3_ultra": 13332.6, "apple_m4": 1953.5, "apple_m4_max": 8888.4, "apple_m4_pro": 4444.2, "nvidia_rtx_3060_12gb": 5860.5, "nvidia_rtx_3090": 15237.2, "nvidia_rtx_4060_ti_16gb": 4688.4, "nvidia_rtx_4070_ti_super": 10939.5, "nvidia_rtx_4080_super": 11981.4, "nvidia_rtx_4090": 16409.3, "nvidia_rtx_5080": 15627.9, "nvidia_rtx_5090": 29172.1}, "model_id": "microsoft/deberta-v3-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "deberta v3 large", "fits": {"amd_rx_7900_xt": 3684.2, "amd_rx_7900_xtx": 4421.1, "amd_rx_9070_xt": 2970.4, "apple_m1_max": 1842.1, "apple_m2_max": 1842.1, "apple_m2_ultra": 3684.2, "apple_m3_max": 1842.1, "apple_m3_ultra": 3771.7, "apple_m4": 552.6, "apple_m4_max": 2514.5, "apple_m4_pro": 1257.2, "nvidia_rtx_3060_12gb": 1657.9, "nvidia_rtx_3090": 4310.5, "nvidia_rtx_4060_ti_16gb": 1326.3, "nvidia_rtx_4070_ti_super": 3094.7, "nvidia_rtx_4080_super": 3389.5, "nvidia_rtx_4090": 4642.1, "nvidia_rtx_5080": 4421.1, "nvidia_rtx_5090": 8252.6}, "model_id": "microsoft/deberta-v3-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "deberta v3 small", "fits": {"amd_rx_7900_xt": 25454.5, "amd_rx_7900_xtx": 30545.5, "amd_rx_9070_xt": 20522.7, "apple_m1_max": 12727.3, "apple_m2_max": 12727.3, "apple_m2_ultra": 25454.5, "apple_m3_max": 12727.3, "apple_m3_ultra": 26059.1, "apple_m4": 3818.2, "apple_m4_max": 17372.7, "apple_m4_pro": 8686.4, "nvidia_rtx_3060_12gb": 11454.5, "nvidia_rtx_3090": 29781.8, "nvidia_rtx_4060_ti_16gb": 9163.6, "nvidia_rtx_4070_ti_super": 21381.8, "nvidia_rtx_4080_super": 23418.2, "nvidia_rtx_4090": 32072.7, "nvidia_rtx_5080": 30545.5, "nvidia_rtx_5090": 57018.2}, "model_id": "microsoft/deberta-v3-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "deberta xlarge mnli", "fits": {"amd_rx_7900_xt": 1493.3, "amd_rx_7900_xtx": 1792.0, "amd_rx_9070_xt": 1204.0, "apple_m1_max": 746.7, "apple_m2_max": 746.7, "apple_m2_ultra": 1493.3, "apple_m3_max": 746.7, "apple_m3_ultra": 1528.8, "apple_m4": 224.0, "apple_m4_max": 1019.2, "apple_m4_pro": 509.6, "nvidia_rtx_3060_12gb": 672.0, "nvidia_rtx_3090": 1747.2, "nvidia_rtx_4060_ti_16gb": 537.6, "nvidia_rtx_4070_ti_super": 1254.4, "nvidia_rtx_4080_super": 1373.9, "nvidia_rtx_4090": 1881.6, "nvidia_rtx_5080": 1792.0, "nvidia_rtx_5090": 3345.1}, "model_id": "microsoft/deberta-xlarge-mnli", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {"bbh": 30.1, "gpqa_diamond": 25.4, "ifeval": 14.8, "math_500": 0.0, "mmlu_pro": 11.2, "musr": 42.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "DialoGPT medium", "fits": {"amd_rx_7900_xt": 3246.4, "amd_rx_7900_xtx": 3895.7, "amd_rx_9070_xt": 2617.4, "apple_m1_max": 1623.2, "apple_m2_max": 1623.2, "apple_m2_ultra": 3246.4, "apple_m3_max": 1623.2, "apple_m3_ultra": 3323.5, "apple_m4": 487.0, "apple_m4_max": 2215.7, "apple_m4_pro": 1107.8, "nvidia_rtx_3060_12gb": 1460.9, "nvidia_rtx_3090": 3798.3, "nvidia_rtx_4060_ti_16gb": 1168.7, "nvidia_rtx_4070_ti_super": 2727.0, "nvidia_rtx_4080_super": 2986.7, "nvidia_rtx_4090": 4090.4, "nvidia_rtx_5080": 3895.7, "nvidia_rtx_5090": 7271.9}, "model_id": "microsoft/dialogpt-medium", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 1024, "cost_input": null, "display_name": "Florence 2 base", "fits": {"amd_rx_7900_xt": 4827.6, "amd_rx_7900_xtx": 5793.1, "amd_rx_9070_xt": 3892.2, "apple_m1_max": 2413.8, "apple_m2_max": 2413.8, "apple_m2_ultra": 4827.6, "apple_m3_max": 2413.8, "apple_m3_ultra": 4942.2, "apple_m4": 724.1, "apple_m4_max": 3294.8, "apple_m4_pro": 1647.4, "nvidia_rtx_3060_12gb": 2172.4, "nvidia_rtx_3090": 5648.3, "nvidia_rtx_4060_ti_16gb": 1737.9, "nvidia_rtx_4070_ti_super": 4055.2, "nvidia_rtx_4080_super": 4441.4, "nvidia_rtx_4090": 6082.8, "nvidia_rtx_5080": 5793.1, "nvidia_rtx_5090": 10813.8}, "model_id": "microsoft/florence-2-base", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 1024, "cost_input": null, "display_name": "Florence 2 large", "fits": {"amd_rx_7900_xt": 1454.5, "amd_rx_7900_xtx": 1745.5, "amd_rx_9070_xt": 1172.7, "apple_m1_max": 727.3, "apple_m2_max": 727.3, "apple_m2_ultra": 1454.5, "apple_m3_max": 727.3, "apple_m3_ultra": 1489.1, "apple_m4": 218.2, "apple_m4_max": 992.7, "apple_m4_pro": 496.4, "nvidia_rtx_3060_12gb": 654.5, "nvidia_rtx_3090": 1701.8, "nvidia_rtx_4060_ti_16gb": 523.6, "nvidia_rtx_4070_ti_super": 1221.8, "nvidia_rtx_4080_super": 1338.2, "nvidia_rtx_4090": 1832.7, "nvidia_rtx_5080": 1745.5, "nvidia_rtx_5090": 3258.2}, "model_id": "microsoft/florence-2-large", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "graphcodebert base", "fits": {"amd_rx_7900_xt": 8960.0, "amd_rx_7900_xtx": 10752.0, "amd_rx_9070_xt": 7224.0, "apple_m1_max": 4480.0, "apple_m2_max": 4480.0, "apple_m2_ultra": 8960.0, "apple_m3_max": 4480.0, "apple_m3_ultra": 9172.8, "apple_m4": 1344.0, "apple_m4_max": 6115.2, "apple_m4_pro": 3057.6, "nvidia_rtx_3060_12gb": 4032.0, "nvidia_rtx_3090": 10483.2, "nvidia_rtx_4060_ti_16gb": 3225.6, "nvidia_rtx_4070_ti_super": 7526.4, "nvidia_rtx_4080_super": 8243.2, "nvidia_rtx_4090": 11289.6, "nvidia_rtx_5080": 10752.0, "nvidia_rtx_5090": 20070.4}, "model_id": "microsoft/graphcodebert-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "kosmos 2 patch14 224", "fits": {"amd_rx_7900_xt": 700.0, "amd_rx_7900_xtx": 840.0, "amd_rx_9070_xt": 564.4, "apple_m1_max": 350.0, "apple_m2_max": 350.0, "apple_m2_ultra": 700.0, "apple_m3_max": 350.0, "apple_m3_ultra": 716.6, "apple_m4": 105.0, "apple_m4_max": 477.7, "apple_m4_pro": 238.9, "nvidia_rtx_3060_12gb": 315.0, "nvidia_rtx_3090": 819.0, "nvidia_rtx_4060_ti_16gb": 252.0, "nvidia_rtx_4070_ti_super": 588.0, "nvidia_rtx_4080_super": 644.0, "nvidia_rtx_4090": 882.0, "nvidia_rtx_5080": 840.0, "nvidia_rtx_5090": 1568.0}, "model_id": "microsoft/kosmos-2-patch14-224", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "layoutlmv2 base uncased", "fits": {"amd_rx_7900_xt": 5600.0, "amd_rx_7900_xtx": 6720.0, "amd_rx_9070_xt": 4515.0, "apple_m1_max": 2800.0, "apple_m2_max": 2800.0, "apple_m2_ultra": 5600.0, "apple_m3_max": 2800.0, "apple_m3_ultra": 5733.0, "apple_m4": 840.0, "apple_m4_max": 3822.0, "apple_m4_pro": 1911.0, "nvidia_rtx_3060_12gb": 2520.0, "nvidia_rtx_3090": 6552.0, "nvidia_rtx_4060_ti_16gb": 2016.0, "nvidia_rtx_4070_ti_super": 4704.0, "nvidia_rtx_4080_super": 5152.0, "nvidia_rtx_4090": 7056.0, "nvidia_rtx_5080": 6720.0, "nvidia_rtx_5090": 12544.0}, "model_id": "microsoft/layoutlmv2-base-uncased", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "layoutlmv3 base", "fits": {"amd_rx_7900_xt": 8421.1, "amd_rx_7900_xtx": 10105.3, "amd_rx_9070_xt": 6789.5, "apple_m1_max": 4210.5, "apple_m2_max": 4210.5, "apple_m2_ultra": 8421.1, "apple_m3_max": 4210.5, "apple_m3_ultra": 8621.1, "apple_m4": 1263.2, "apple_m4_max": 5747.4, "apple_m4_pro": 2873.7, "nvidia_rtx_3060_12gb": 3789.5, "nvidia_rtx_3090": 9852.6, "nvidia_rtx_4060_ti_16gb": 3031.6, "nvidia_rtx_4070_ti_super": 7073.7, "nvidia_rtx_4080_super": 7747.4, "nvidia_rtx_4090": 10610.5, "nvidia_rtx_5080": 10105.3, "nvidia_rtx_5090": 18863.2}, "model_id": "microsoft/layoutlmv3-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "llmlingua 2 xlm roberta large meetingbank", "fits": {"amd_rx_7900_xt": 3154.9, "amd_rx_7900_xtx": 3785.9, "amd_rx_9070_xt": 2543.7, "apple_m1_max": 1577.5, "apple_m2_max": 1577.5, "apple_m2_ultra": 3154.9, "apple_m3_max": 1577.5, "apple_m3_ultra": 3229.9, "apple_m4": 473.2, "apple_m4_max": 2153.2, "apple_m4_pro": 1076.6, "nvidia_rtx_3060_12gb": 1419.7, "nvidia_rtx_3090": 3691.3, "nvidia_rtx_4060_ti_16gb": 1135.8, "nvidia_rtx_4070_ti_super": 2650.1, "nvidia_rtx_4080_super": 2902.5, "nvidia_rtx_4090": 3975.2, "nvidia_rtx_5080": 3785.9, "nvidia_rtx_5090": 7067.0}, "model_id": "microsoft/llmlingua-2-xlm-roberta-large-meetingbank", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "markuplm base", "fits": {"amd_rx_7900_xt": 10275.2, "amd_rx_7900_xtx": 12330.3, "amd_rx_9070_xt": 8284.4, "apple_m1_max": 5137.6, "apple_m2_max": 5137.6, "apple_m2_ultra": 10275.2, "apple_m3_max": 5137.6, "apple_m3_ultra": 10519.3, "apple_m4": 1541.3, "apple_m4_max": 7012.8, "apple_m4_pro": 3506.4, "nvidia_rtx_3060_12gb": 4623.9, "nvidia_rtx_3090": 12022.0, "nvidia_rtx_4060_ti_16gb": 3699.1, "nvidia_rtx_4070_ti_super": 8631.2, "nvidia_rtx_4080_super": 9453.2, "nvidia_rtx_4090": 12946.8, "nvidia_rtx_5080": 12330.3, "nvidia_rtx_5090": 23016.5}, "model_id": "microsoft/markuplm-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "mdeberta v3 base", "fits": {"amd_rx_7900_xt": 4014.3, "amd_rx_7900_xtx": 4817.2, "amd_rx_9070_xt": 3236.6, "apple_m1_max": 2007.2, "apple_m2_max": 2007.2, "apple_m2_ultra": 4014.3, "apple_m3_max": 2007.2, "apple_m3_ultra": 4109.7, "apple_m4": 602.2, "apple_m4_max": 2739.8, "apple_m4_pro": 1369.9, "nvidia_rtx_3060_12gb": 1806.5, "nvidia_rtx_3090": 4696.8, "nvidia_rtx_4060_ti_16gb": 1445.2, "nvidia_rtx_4070_ti_super": 3372.0, "nvidia_rtx_4080_super": 3693.2, "nvidia_rtx_4090": 5058.1, "nvidia_rtx_5080": 4817.2, "nvidia_rtx_5090": 8992.1}, "model_id": "microsoft/mdeberta-v3-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 61.0, "bbh": 48.8, "gpqa_diamond": 27.2, "gsm8k": 55.0, "hellaswag": 74.9, "ifeval": 27.4, "math_500": 2.9, "mmlu_abstract_algebra": 29.0, "mmlu_anatomy": 44.4, "mmlu_astronomy": 58.6, "mmlu_business_ethics": 56.0, "mmlu_clinical_knowledge": 60.4, "mmlu_college_biology": 66.7, "mmlu_college_chemistry": 40.0, "mmlu_college_computer_science": 41.0, "mmlu_college_mathematics": 38.0, "mmlu_college_medicine": 59.0, "mmlu_college_physics": 37.3, "mmlu_computer_security": 74.0, "mmlu_conceptual_physics": 52.3, "mmlu_econometrics": 38.6, "mmlu_electrical_engineering": 55.2, "mmlu_elementary_mathematics": 46.0, "mmlu_formal_logic": 35.7, "mmlu_global_facts": 38.0, "mmlu_high_school_biology": 67.1, "mmlu_high_school_chemistry": 47.3, "mmlu_high_school_computer_science": 64.0, "mmlu_high_school_european_history": 64.8, "mmlu_high_school_geography": 75.3, "mmlu_high_school_government_and_politics": 80.8, "mmlu_high_school_macroeconomics": 58.2, "mmlu_high_school_mathematics": 33.0, "mmlu_high_school_microeconomics": 62.2, "mmlu_high_school_physics": 38.4, "mmlu_high_school_psychology": 79.3, "mmlu_high_school_statistics": 47.2, "mmlu_high_school_us_history": 66.7, "mmlu_high_school_world_history": 74.3, "mmlu_human_aging": 65.9, "mmlu_human_sexuality": 70.2, "mmlu_international_law": 72.7, "mmlu_jurisprudence": 73.1, "mmlu_logical_fallacies": 73.0, "mmlu_machine_learning": 48.2, "mmlu_management": 70.9, "mmlu_marketing": 82.5, "mmlu_medical_genetics": 64.0, "mmlu_miscellaneous": 69.2, "mmlu_moral_disputes": 66.5, "mmlu_moral_scenarios": 29.9, "mmlu_nutrition": 62.1, "mmlu_philosophy": 62.1, "mmlu_prehistory": 62.0, "mmlu_pro": 26.3, "mmlu_professional_accounting": 44.0, "mmlu_professional_law": 43.7, "mmlu_professional_medicine": 48.2, "mmlu_professional_psychology": 56.4, "mmlu_public_relations": 62.7, "mmlu_security_studies": 71.8, "mmlu_sociology": 80.1, "mmlu_us_foreign_policy": 77.0, "mmlu_virology": 47.0, "mmlu_world_religions": 69.6, "musr": 41.0, "truthfulqa": 44.2, "winogrande": 73.5}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "phi 2", "fits": {"amd_rx_7900_xt": 414.8, "amd_rx_7900_xtx": 497.8, "amd_rx_9070_xt": 334.4, "apple_m1_max": 207.4, "apple_m2_max": 207.4, "apple_m2_ultra": 414.8, "apple_m3_max": 207.4, "apple_m3_ultra": 424.7, "apple_m4": 62.2, "apple_m4_max": 283.1, "apple_m4_pro": 141.6, "nvidia_rtx_3060_12gb": 186.7, "nvidia_rtx_3090": 485.3, "nvidia_rtx_4060_ti_16gb": 149.3, "nvidia_rtx_4070_ti_super": 348.4, "nvidia_rtx_4080_super": 381.6, "nvidia_rtx_4090": 522.7, "nvidia_rtx_5080": 497.8, "nvidia_rtx_5090": 929.2}, "model_id": "microsoft/phi-2", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2024-07"}, {"benchmark_scores": {"artificial_analysis_quality_index": 58.0, "artificial_analysis_speed_index": 92.0, "bbh": 38.1, "gpqa_diamond": 28.5, "gsm8k": 86.2, "ifeval": 72.4, "math_500": 52.8, "mgsm": 47.9, "mmlu_pro": 42.3, "musr": 11.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Phi 3.5 mini instruct", "fits": {"amd_rx_7900_xt": 294.7, "amd_rx_7900_xtx": 353.7, "amd_rx_9070_xt": 237.6, "apple_m1_max": 147.4, "apple_m2_max": 147.4, "apple_m2_ultra": 294.7, "apple_m3_max": 147.4, "apple_m3_ultra": 301.7, "apple_m4": 44.2, "apple_m4_max": 201.2, "apple_m4_pro": 100.6, "nvidia_rtx_3060_12gb": 132.6, "nvidia_rtx_3090": 344.8, "nvidia_rtx_4060_ti_16gb": 106.1, "nvidia_rtx_4070_ti_super": 247.6, "nvidia_rtx_4080_super": 271.2, "nvidia_rtx_4090": 371.4, "nvidia_rtx_5080": 353.7, "nvidia_rtx_5090": 660.2}, "model_id": "microsoft/phi-3-5-mini-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 78.1, "chartqa": 81.8, "mathvista": 43.9, "mmmu": 43.0}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Phi 3.5 vision instruct", "fits": {"amd_rx_7900_xt": 266.7, "amd_rx_7900_xtx": 320.0, "amd_rx_9070_xt": 215.0, "apple_m1_max": 133.3, "apple_m2_max": 133.3, "apple_m2_ultra": 266.7, "apple_m3_max": 133.3, "apple_m3_ultra": 273.0, "apple_m4": 40.0, "apple_m4_max": 182.0, "apple_m4_pro": 91.0, "nvidia_rtx_3060_12gb": 120.0, "nvidia_rtx_3090": 312.0, "nvidia_rtx_4060_ti_16gb": 96.0, "nvidia_rtx_4070_ti_super": 224.0, "nvidia_rtx_4080_super": 245.3, "nvidia_rtx_4090": 336.0, "nvidia_rtx_5080": 320.0, "nvidia_rtx_5090": 597.3}, "model_id": "microsoft/phi-3-5-vision-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 63.1, "bbh": 55.7, "gpqa_diamond": 31.8, "gsm8k": 69.5, "hellaswag": 80.1, "ifeval": 59.8, "math_500": 14.0, "mmlu_abstract_algebra": 38.0, "mmlu_anatomy": 65.2, "mmlu_astronomy": 78.9, "mmlu_business_ethics": 68.0, "mmlu_clinical_knowledge": 74.0, "mmlu_college_biology": 79.9, "mmlu_college_chemistry": 44.0, "mmlu_college_computer_science": 56.0, "mmlu_college_mathematics": 40.0, "mmlu_college_medicine": 68.2, "mmlu_college_physics": 43.1, "mmlu_computer_security": 76.0, "mmlu_conceptual_physics": 69.8, "mmlu_econometrics": 49.1, "mmlu_electrical_engineering": 62.1, "mmlu_elementary_mathematics": 48.9, "mmlu_formal_logic": 57.1, "mmlu_global_facts": 37.0, "mmlu_high_school_biology": 83.5, "mmlu_high_school_chemistry": 60.1, "mmlu_high_school_computer_science": 67.0, "mmlu_high_school_european_history": 81.2, "mmlu_high_school_geography": 84.8, "mmlu_high_school_government_and_politics": 88.6, "mmlu_high_school_macroeconomics": 73.3, "mmlu_high_school_mathematics": 40.0, "mmlu_high_school_microeconomics": 82.8, "mmlu_high_school_physics": 44.4, "mmlu_high_school_psychology": 89.0, "mmlu_high_school_statistics": 64.8, "mmlu_high_school_us_history": 83.3, "mmlu_high_school_world_history": 79.7, "mmlu_human_aging": 70.9, "mmlu_human_sexuality": 75.6, "mmlu_international_law": 82.6, "mmlu_jurisprudence": 77.8, "mmlu_logical_fallacies": 81.6, "mmlu_machine_learning": 54.5, "mmlu_management": 82.5, "mmlu_marketing": 88.9, "mmlu_medical_genetics": 83.0, "mmlu_miscellaneous": 81.6, "mmlu_moral_disputes": 74.0, "mmlu_moral_scenarios": 58.4, "mmlu_nutrition": 77.8, "mmlu_philosophy": 73.6, "mmlu_prehistory": 77.2, "mmlu_pro": 37.3, "mmlu_professional_accounting": 58.5, "mmlu_professional_law": 49.7, "mmlu_professional_medicine": 72.4, "mmlu_professional_psychology": 75.3, "mmlu_public_relations": 67.3, "mmlu_security_studies": 74.7, "mmlu_sociology": 85.6, "mmlu_us_foreign_policy": 87.0, "mmlu_virology": 47.0, "mmlu_world_religions": 80.7, "musr": 39.4, "truthfulqa": 54.1, "winogrande": 72.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Phi 3 mini 128K instruct", "fits": {"amd_rx_7900_xt": 294.7, "amd_rx_7900_xtx": 353.7, "amd_rx_9070_xt": 237.6, "apple_m1_max": 147.4, "apple_m2_max": 147.4, "apple_m2_ultra": 294.7, "apple_m3_max": 147.4, "apple_m3_ultra": 301.7, "apple_m4": 44.2, "apple_m4_max": 201.2, "apple_m4_pro": 100.6, "nvidia_rtx_3060_12gb": 132.6, "nvidia_rtx_3090": 344.8, "nvidia_rtx_4060_ti_16gb": 106.1, "nvidia_rtx_4070_ti_super": 247.6, "nvidia_rtx_4080_super": 271.2, "nvidia_rtx_4090": 371.4, "nvidia_rtx_5080": 353.7, "nvidia_rtx_5090": 660.2}, "model_id": "microsoft/phi-3-mini-128k-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 63.0, "bbh": 55.0, "gpqa_diamond": 32.3, "gsm8k": 74.5, "hellaswag": 80.6, "ifeval": 54.4, "math_500": 16.4, "mmlu_abstract_algebra": 36.0, "mmlu_anatomy": 66.7, "mmlu_astronomy": 77.0, "mmlu_business_ethics": 68.0, "mmlu_clinical_knowledge": 74.3, "mmlu_college_biology": 82.6, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 55.0, "mmlu_college_mathematics": 35.0, "mmlu_college_medicine": 68.8, "mmlu_college_physics": 36.3, "mmlu_computer_security": 78.0, "mmlu_conceptual_physics": 70.6, "mmlu_econometrics": 45.6, "mmlu_electrical_engineering": 57.9, "mmlu_elementary_mathematics": 52.6, "mmlu_formal_logic": 58.7, "mmlu_global_facts": 40.0, "mmlu_high_school_biology": 83.9, "mmlu_high_school_chemistry": 60.1, "mmlu_high_school_computer_science": 73.0, "mmlu_high_school_european_history": 80.6, "mmlu_high_school_geography": 86.9, "mmlu_high_school_government_and_politics": 90.7, "mmlu_high_school_macroeconomics": 75.4, "mmlu_high_school_mathematics": 39.6, "mmlu_high_school_microeconomics": 83.6, "mmlu_high_school_physics": 44.4, "mmlu_high_school_psychology": 88.3, "mmlu_high_school_statistics": 61.6, "mmlu_high_school_us_history": 79.9, "mmlu_high_school_world_history": 79.3, "mmlu_human_aging": 69.1, "mmlu_human_sexuality": 76.3, "mmlu_international_law": 85.1, "mmlu_jurisprudence": 79.6, "mmlu_logical_fallacies": 80.4, "mmlu_machine_learning": 55.4, "mmlu_management": 80.6, "mmlu_marketing": 90.2, "mmlu_medical_genetics": 81.0, "mmlu_miscellaneous": 82.8, "mmlu_moral_disputes": 75.4, "mmlu_moral_scenarios": 58.5, "mmlu_nutrition": 75.8, "mmlu_philosophy": 76.2, "mmlu_prehistory": 78.1, "mmlu_pro": 40.3, "mmlu_professional_accounting": 58.2, "mmlu_professional_law": 51.3, "mmlu_professional_medicine": 76.1, "mmlu_professional_psychology": 75.8, "mmlu_public_relations": 73.6, "mmlu_security_studies": 76.7, "mmlu_sociology": 86.1, "mmlu_us_foreign_policy": 85.0, "mmlu_virology": 49.4, "mmlu_world_religions": 83.6, "musr": 42.8, "truthfulqa": 59.9, "winogrande": 72.4}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": 0.1, "display_name": "Phi 3 mini 4K instruct", "fits": {"amd_rx_7900_xt": 294.7, "amd_rx_7900_xtx": 353.7, "amd_rx_9070_xt": 237.6, "apple_m1_max": 147.4, "apple_m2_max": 147.4, "apple_m2_ultra": 294.7, "apple_m3_max": 147.4, "apple_m3_ultra": 301.7, "apple_m4": 44.2, "apple_m4_max": 201.2, "apple_m4_pro": 100.6, "nvidia_rtx_3060_12gb": 132.6, "nvidia_rtx_3090": 344.8, "nvidia_rtx_4060_ti_16gb": 106.1, "nvidia_rtx_4070_ti_super": 247.6, "nvidia_rtx_4080_super": 271.2, "nvidia_rtx_4090": 371.4, "nvidia_rtx_5080": 353.7, "nvidia_rtx_5090": 660.2}, "model_id": "microsoft/phi-3-mini-4k-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2024-07"}, {"benchmark_scores": {"alpaca_eval": 32.5, "arena_elo_coding": 1210.0, "arena_elo_math": 1220.0, "arena_elo_overall": 1200.0, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 88.0, "bbh": 56.9, "gpqa_diamond": 42.1, "gsm8k": 88.6, "humaneval": 70.2, "ifeval": 72.0, "live_code_bench": 23.1, "math_500": 72.5, "mgsm": 63.9, "mmlu_astronomy": 68.8, "mmlu_biology": 77.8, "mmlu_business_ethics": 70.2, "mmlu_chemistry": 70.5, "mmlu_clinical_knowledge": 76.2, "mmlu_computer_science": 78.5, "mmlu_jurisprudence": 66.8, "mmlu_physics": 74.2, "mmlu_pro": 58.2, "mmlu_professional_accounting": 61.2, "mmlu_professional_law": 65.5, "mt_bench": 8.2, "multipl_e": 72.0, "multipl_e_cpp": 72.1, "multipl_e_csharp": 62.5, "multipl_e_go": 67.5, "multipl_e_java": 76.5, "multipl_e_javascript": 74.2, "multipl_e_julia": 40.2, "multipl_e_kotlin": 50.5, "multipl_e_lua": 35.2, "multipl_e_perl": 30.5, "multipl_e_php": 58.8, "multipl_e_python": 79.2, "multipl_e_r": 36.5, "multipl_e_ruby": 42.5, "multipl_e_rust": 60.8, "multipl_e_scala": 38.2, "multipl_e_swift": 44.2, "multipl_e_typescript": 73.8, "musr": 38.7, "wildbench": 58.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-2"}, "context_window": 131072, "cost_input": 0.04, "display_name": "Phi 4 mini instruct", "fits": {"amd_rx_7900_xt": 294.7, "amd_rx_7900_xtx": 353.7, "amd_rx_9070_xt": 237.6, "apple_m1_max": 147.4, "apple_m2_max": 147.4, "apple_m2_ultra": 294.7, "apple_m3_max": 147.4, "apple_m3_ultra": 301.7, "apple_m4": 44.2, "apple_m4_max": 201.2, "apple_m4_pro": 100.6, "nvidia_rtx_3060_12gb": 132.6, "nvidia_rtx_3090": 344.8, "nvidia_rtx_4060_ti_16gb": 106.1, "nvidia_rtx_4070_ti_super": 247.6, "nvidia_rtx_4080_super": 271.2, "nvidia_rtx_4090": 371.4, "nvidia_rtx_5080": 353.7, "nvidia_rtx_5090": 660.2}, "model_id": "microsoft/phi-4-mini-instruct", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 32.5, "arena_elo_coding": 1250.0, "arena_elo_math": 1260.0, "arena_elo_overall": 1240.0, "bbh": 54.3, "gpqa_diamond": 52.5, "humaneval": 78.5, "ifeval": 78.0, "math_500": 82.5, "mmlu_astronomy": 68.8, "mmlu_biology": 77.8, "mmlu_business_ethics": 70.2, "mmlu_chemistry": 70.5, "mmlu_clinical_knowledge": 76.2, "mmlu_computer_science": 78.5, "mmlu_jurisprudence": 66.8, "mmlu_physics": 74.2, "mmlu_pro": 68.5, "mmlu_professional_accounting": 61.2, "mmlu_professional_law": 65.5, "mt_bench": 8.2, "musr": 23.5, "wildbench": 58.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2"}, "context_window": 131072, "cost_input": 0.07, "display_name": "Phi 4 multimodal instruct", "fits": {"amd_rx_7900_xt": 200.0, "amd_rx_7900_xtx": 240.0, "amd_rx_9070_xt": 161.2, "apple_m1_max": 100.0, "apple_m2_max": 100.0, "apple_m2_ultra": 200.0, "apple_m3_max": 100.0, "apple_m3_ultra": 204.8, "apple_m4": 30.0, "apple_m4_max": 136.5, "apple_m4_pro": 68.2, "nvidia_rtx_3060_12gb": 90.0, "nvidia_rtx_3090": 234.0, "nvidia_rtx_4060_ti_16gb": 72.0, "nvidia_rtx_4070_ti_super": 168.0, "nvidia_rtx_4080_super": 184.0, "nvidia_rtx_4090": 252.0, "nvidia_rtx_5080": 240.0, "nvidia_rtx_5090": 448.0}, "model_id": "microsoft/phi-4-multimodal-instruct", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "audio-asr", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 32.5, "arena_elo_coding": 1250.0, "arena_elo_hard_prompts": 1277.2, "arena_elo_math": 1260.0, "arena_elo_overall": 1240.0, "arena_elo_style_control": 1255.4, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 88.0, "bbh": 54.3, "gpqa_diamond": 52.5, "gsm8k": 93.5, "humaneval": 78.5, "ifeval": 78.0, "live_code_bench": 23.1, "math_500": 82.5, "medqa": 77.8, "mgsm": 80.6, "mmlu_astronomy": 68.8, "mmlu_biology": 77.8, "mmlu_business_ethics": 70.2, "mmlu_chemistry": 70.5, "mmlu_clinical_knowledge": 76.2, "mmlu_computer_science": 78.5, "mmlu_jurisprudence": 66.8, "mmlu_physics": 74.2, "mmlu_pro": 68.5, "mmlu_professional_accounting": 61.2, "mmlu_professional_law": 65.5, "mt_bench": 8.2, "multipl_e": 72.0, "multipl_e_cpp": 72.1, "multipl_e_csharp": 72.8, "multipl_e_go": 67.5, "multipl_e_java": 76.5, "multipl_e_javascript": 74.2, "multipl_e_julia": 52.5, "multipl_e_kotlin": 62.8, "multipl_e_lua": 45.5, "multipl_e_perl": 40.8, "multipl_e_php": 69.5, "multipl_e_python": 79.2, "multipl_e_r": 48.2, "multipl_e_ruby": 55.2, "multipl_e_rust": 60.8, "multipl_e_scala": 50.5, "multipl_e_swift": 55.8, "multipl_e_typescript": 73.8, "musr": 23.5, "wildbench": 58.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-2"}, "context_window": 16384, "cost_input": 0.07, "display_name": "phi 4", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "microsoft/phi-4", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "Microsoft", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Phi tiny MoE instruct", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "microsoft/phi-tiny-moe-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "resnet 50", "fits": {"amd_rx_7900_xt": 43750.0, "amd_rx_7900_xtx": 52500.0, "amd_rx_9070_xt": 35273.4, "apple_m1_max": 21875.0, "apple_m2_max": 21875.0, "apple_m2_ultra": 43750.0, "apple_m3_max": 21875.0, "apple_m3_ultra": 44789.1, "apple_m4": 6562.5, "apple_m4_max": 29859.4, "apple_m4_pro": 14929.7, "nvidia_rtx_3060_12gb": 19687.5, "nvidia_rtx_3090": 51187.5, "nvidia_rtx_4060_ti_16gb": 15750.0, "nvidia_rtx_4070_ti_super": 36750.0, "nvidia_rtx_4080_super": 40250.0, "nvidia_rtx_4090": 55125.0, "nvidia_rtx_5080": 52500.0, "nvidia_rtx_5090": 98000.0}, "model_id": "microsoft/resnet-50", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "speecht5 tts", "fits": {"amd_rx_7900_xt": 7777.8, "amd_rx_7900_xtx": 9333.3, "amd_rx_9070_xt": 6270.8, "apple_m1_max": 3888.9, "apple_m2_max": 3888.9, "apple_m2_ultra": 7777.8, "apple_m3_max": 3888.9, "apple_m3_ultra": 7962.5, "apple_m4": 1166.7, "apple_m4_max": 5308.3, "apple_m4_pro": 2654.2, "nvidia_rtx_3060_12gb": 3500.0, "nvidia_rtx_3090": 9100.0, "nvidia_rtx_4060_ti_16gb": 2800.0, "nvidia_rtx_4070_ti_super": 6533.3, "nvidia_rtx_4080_super": 7155.6, "nvidia_rtx_4090": 9800.0, "nvidia_rtx_5080": 9333.3, "nvidia_rtx_5090": 17422.2}, "model_id": "microsoft/speecht5-tts", "model_subtypes": [], "model_type": "audio-tts", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "swinv2 tiny patch4 window16 256", "fits": {"amd_rx_7900_xt": 40000.0, "amd_rx_7900_xtx": 48000.0, "amd_rx_9070_xt": 32250.0, "apple_m1_max": 20000.0, "apple_m2_max": 20000.0, "apple_m2_ultra": 40000.0, "apple_m3_max": 20000.0, "apple_m3_ultra": 40950.0, "apple_m4": 6000.0, "apple_m4_max": 27300.0, "apple_m4_pro": 13650.0, "nvidia_rtx_3060_12gb": 18000.0, "nvidia_rtx_3090": 46800.0, "nvidia_rtx_4060_ti_16gb": 14400.0, "nvidia_rtx_4070_ti_super": 33600.0, "nvidia_rtx_4080_super": 36800.0, "nvidia_rtx_4090": 50400.0, "nvidia_rtx_5080": 48000.0, "nvidia_rtx_5090": 89600.0}, "model_id": "microsoft/swinv2-tiny-patch4-window16-256", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "table transformer detection", "fits": {"amd_rx_7900_xt": 48695.7, "amd_rx_7900_xtx": 58434.8, "amd_rx_9070_xt": 39260.9, "apple_m1_max": 24347.8, "apple_m2_max": 24347.8, "apple_m2_ultra": 48695.7, "apple_m3_max": 24347.8, "apple_m3_ultra": 49852.2, "apple_m4": 7304.3, "apple_m4_max": 33234.8, "apple_m4_pro": 16617.4, "nvidia_rtx_3060_12gb": 21913.0, "nvidia_rtx_3090": 56973.9, "nvidia_rtx_4060_ti_16gb": 17530.4, "nvidia_rtx_4070_ti_super": 40904.3, "nvidia_rtx_4080_super": 44800.0, "nvidia_rtx_4090": 61356.5, "nvidia_rtx_5080": 58434.8, "nvidia_rtx_5090": 109078.3}, "model_id": "microsoft/table-transformer-detection", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "table transformer structure recognition v1.1 all", "fits": {"amd_rx_7900_xt": 48695.7, "amd_rx_7900_xtx": 58434.8, "amd_rx_9070_xt": 39260.9, "apple_m1_max": 24347.8, "apple_m2_max": 24347.8, "apple_m2_ultra": 48695.7, "apple_m3_max": 24347.8, "apple_m3_ultra": 49852.2, "apple_m4": 7304.3, "apple_m4_max": 33234.8, "apple_m4_pro": 16617.4, "nvidia_rtx_3060_12gb": 21913.0, "nvidia_rtx_3090": 56973.9, "nvidia_rtx_4060_ti_16gb": 17530.4, "nvidia_rtx_4070_ti_super": 40904.3, "nvidia_rtx_4080_super": 44800.0, "nvidia_rtx_4090": 61356.5, "nvidia_rtx_5080": 58434.8, "nvidia_rtx_5090": 109078.3}, "model_id": "microsoft/table-transformer-structure-recognition-v1-1-all", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "table transformer structure recognition", "fits": {"amd_rx_7900_xt": 48695.7, "amd_rx_7900_xtx": 58434.8, "amd_rx_9070_xt": 39260.9, "apple_m1_max": 24347.8, "apple_m2_max": 24347.8, "apple_m2_ultra": 48695.7, "apple_m3_max": 24347.8, "apple_m3_ultra": 49852.2, "apple_m4": 7304.3, "apple_m4_max": 33234.8, "apple_m4_pro": 16617.4, "nvidia_rtx_3060_12gb": 21913.0, "nvidia_rtx_3090": 56973.9, "nvidia_rtx_4060_ti_16gb": 17530.4, "nvidia_rtx_4070_ti_super": 40904.3, "nvidia_rtx_4080_super": 44800.0, "nvidia_rtx_4090": 61356.5, "nvidia_rtx_5080": 58434.8, "nvidia_rtx_5090": 109078.3}, "model_id": "microsoft/table-transformer-structure-recognition", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1024, "cost_input": null, "display_name": "tapex base finetuned wikisql", "fits": {"amd_rx_7900_xt": 8000.0, "amd_rx_7900_xtx": 9600.0, "amd_rx_9070_xt": 6450.0, "apple_m1_max": 4000.0, "apple_m2_max": 4000.0, "apple_m2_ultra": 8000.0, "apple_m3_max": 4000.0, "apple_m3_ultra": 8190.0, "apple_m4": 1200.0, "apple_m4_max": 5460.0, "apple_m4_pro": 2730.0, "nvidia_rtx_3060_12gb": 3600.0, "nvidia_rtx_3090": 9360.0, "nvidia_rtx_4060_ti_16gb": 2880.0, "nvidia_rtx_4070_ti_super": 6720.0, "nvidia_rtx_4080_super": 7360.0, "nvidia_rtx_4090": 10080.0, "nvidia_rtx_5080": 9600.0, "nvidia_rtx_5090": 17920.0}, "model_id": "microsoft/tapex-base-finetuned-wikisql", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "TRELLIS image large", "fits": {}, "model_id": "microsoft/trellis-image-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "trocr base printed", "fits": {"amd_rx_7900_xt": 3353.3, "amd_rx_7900_xtx": 4024.0, "amd_rx_9070_xt": 2703.6, "apple_m1_max": 1676.6, "apple_m2_max": 1676.6, "apple_m2_ultra": 3353.3, "apple_m3_max": 1676.6, "apple_m3_ultra": 3432.9, "apple_m4": 503.0, "apple_m4_max": 2288.6, "apple_m4_pro": 1144.3, "nvidia_rtx_3060_12gb": 1509.0, "nvidia_rtx_3090": 3923.4, "nvidia_rtx_4060_ti_16gb": 1207.2, "nvidia_rtx_4070_ti_super": 2816.8, "nvidia_rtx_4080_super": 3085.0, "nvidia_rtx_4090": 4225.1, "nvidia_rtx_5080": 4024.0, "nvidia_rtx_5090": 7511.4}, "model_id": "microsoft/trocr-base-printed", "model_subtypes": [], "model_type": "document-ocr", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "trocr large handwritten", "fits": {"amd_rx_7900_xt": 2007.2, "amd_rx_7900_xtx": 2408.6, "amd_rx_9070_xt": 1618.3, "apple_m1_max": 1003.6, "apple_m2_max": 1003.6, "apple_m2_ultra": 2007.2, "apple_m3_max": 1003.6, "apple_m3_ultra": 2054.8, "apple_m4": 301.1, "apple_m4_max": 1369.9, "apple_m4_pro": 684.9, "nvidia_rtx_3060_12gb": 903.2, "nvidia_rtx_3090": 2348.4, "nvidia_rtx_4060_ti_16gb": 722.6, "nvidia_rtx_4070_ti_super": 1686.0, "nvidia_rtx_4080_super": 1846.6, "nvidia_rtx_4090": 2529.0, "nvidia_rtx_5080": 2408.6, "nvidia_rtx_5090": 4496.1}, "model_id": "microsoft/trocr-large-handwritten", "model_subtypes": [], "model_type": "document-ocr", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "trocr large printed", "fits": {"amd_rx_7900_xt": 2007.2, "amd_rx_7900_xtx": 2408.6, "amd_rx_9070_xt": 1618.3, "apple_m1_max": 1003.6, "apple_m2_max": 1003.6, "apple_m2_ultra": 2007.2, "apple_m3_max": 1003.6, "apple_m3_ultra": 2054.8, "apple_m4": 301.1, "apple_m4_max": 1369.9, "apple_m4_pro": 684.9, "nvidia_rtx_3060_12gb": 903.2, "nvidia_rtx_3090": 2348.4, "nvidia_rtx_4060_ti_16gb": 722.6, "nvidia_rtx_4070_ti_super": 1686.0, "nvidia_rtx_4080_super": 1846.6, "nvidia_rtx_4090": 2529.0, "nvidia_rtx_5080": 2408.6, "nvidia_rtx_5090": 4496.1}, "model_id": "microsoft/trocr-large-printed", "model_subtypes": [], "model_type": "document-ocr", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "unixcoder base", "fits": {"amd_rx_7900_xt": 8888.9, "amd_rx_7900_xtx": 10666.7, "amd_rx_9070_xt": 7166.7, "apple_m1_max": 4444.4, "apple_m2_max": 4444.4, "apple_m2_ultra": 8888.9, "apple_m3_max": 4444.4, "apple_m3_ultra": 9100.0, "apple_m4": 1333.3, "apple_m4_max": 6066.7, "apple_m4_pro": 3033.3, "nvidia_rtx_3060_12gb": 4000.0, "nvidia_rtx_3090": 10400.0, "nvidia_rtx_4060_ti_16gb": 3200.0, "nvidia_rtx_4070_ti_super": 7466.7, "nvidia_rtx_4080_super": 8177.8, "nvidia_rtx_4090": 11200.0, "nvidia_rtx_5080": 10666.7, "nvidia_rtx_5090": 19911.1}, "model_id": "microsoft/unixcoder-base", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "VibeVoice ASR HF", "fits": {}, "model_id": "microsoft/vibevoice-asr-hf", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "VibeVoice ASR", "fits": {}, "model_id": "microsoft/vibevoice-asr", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "VibeVoice Realtime 0.5B", "fits": {"amd_rx_7900_xt": 2240.0, "amd_rx_7900_xtx": 2688.0, "amd_rx_9070_xt": 1806.0, "apple_m1_max": 1120.0, "apple_m2_max": 1120.0, "apple_m2_ultra": 2240.0, "apple_m3_max": 1120.0, "apple_m3_ultra": 2293.2, "apple_m4": 336.0, "apple_m4_max": 1528.8, "apple_m4_pro": 764.4, "nvidia_rtx_3060_12gb": 1008.0, "nvidia_rtx_3090": 2620.8, "nvidia_rtx_4060_ti_16gb": 806.4, "nvidia_rtx_4070_ti_super": 1881.6, "nvidia_rtx_4080_super": 2060.8, "nvidia_rtx_4090": 2822.4, "nvidia_rtx_5080": 2688.0, "nvidia_rtx_5090": 5017.6}, "model_id": "microsoft/vibevoice-realtime-0-5b", "model_subtypes": [], "model_type": "audio-tts", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "wavlm base plus sv", "fits": {"amd_rx_7900_xt": 11789.5, "amd_rx_7900_xtx": 14147.4, "amd_rx_9070_xt": 9505.3, "apple_m1_max": 5894.7, "apple_m2_max": 5894.7, "apple_m2_ultra": 11789.5, "apple_m3_max": 5894.7, "apple_m3_ultra": 12069.5, "apple_m4": 1768.4, "apple_m4_max": 8046.3, "apple_m4_pro": 4023.2, "nvidia_rtx_3060_12gb": 5305.3, "nvidia_rtx_3090": 13793.7, "nvidia_rtx_4060_ti_16gb": 4244.2, "nvidia_rtx_4070_ti_super": 9903.2, "nvidia_rtx_4080_super": 10846.3, "nvidia_rtx_4090": 14854.7, "nvidia_rtx_5080": 14147.4, "nvidia_rtx_5090": 26408.4}, "model_id": "microsoft/wavlm-base-plus-sv", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "wavlm base plus", "fits": {"amd_rx_7900_xt": 11789.5, "amd_rx_7900_xtx": 14147.4, "amd_rx_9070_xt": 9505.3, "apple_m1_max": 5894.7, "apple_m2_max": 5894.7, "apple_m2_ultra": 11789.5, "apple_m3_max": 5894.7, "apple_m3_ultra": 12069.5, "apple_m4": 1768.4, "apple_m4_max": 8046.3, "apple_m4_pro": 4023.2, "nvidia_rtx_3060_12gb": 5305.3, "nvidia_rtx_3090": 13793.7, "nvidia_rtx_4060_ti_16gb": 4244.2, "nvidia_rtx_4070_ti_super": 9903.2, "nvidia_rtx_4080_super": 10846.3, "nvidia_rtx_4090": 14854.7, "nvidia_rtx_5080": 14147.4, "nvidia_rtx_5090": 26408.4}, "model_id": "microsoft/wavlm-base-plus", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "wavlm large", "fits": {"amd_rx_7900_xt": 3533.1, "amd_rx_7900_xtx": 4239.7, "amd_rx_9070_xt": 2848.6, "apple_m1_max": 1766.6, "apple_m2_max": 1766.6, "apple_m2_ultra": 3533.1, "apple_m3_max": 1766.6, "apple_m3_ultra": 3617.0, "apple_m4": 530.0, "apple_m4_max": 2411.4, "apple_m4_pro": 1205.7, "nvidia_rtx_3060_12gb": 1589.9, "nvidia_rtx_3090": 4133.8, "nvidia_rtx_4060_ti_16gb": 1271.9, "nvidia_rtx_4070_ti_super": 2967.8, "nvidia_rtx_4080_super": 3250.5, "nvidia_rtx_4090": 4451.7, "nvidia_rtx_5080": 4239.7, "nvidia_rtx_5090": 7914.2}, "model_id": "microsoft/wavlm-large", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "xclip base patch32", "fits": {"amd_rx_7900_xt": 8484.8, "amd_rx_7900_xtx": 10181.8, "amd_rx_9070_xt": 6840.9, "apple_m1_max": 4242.4, "apple_m2_max": 4242.4, "apple_m2_ultra": 8484.8, "apple_m3_max": 4242.4, "apple_m3_ultra": 8686.4, "apple_m4": 1272.7, "apple_m4_max": 5790.9, "apple_m4_pro": 2895.5, "nvidia_rtx_3060_12gb": 3818.2, "nvidia_rtx_3090": 9927.3, "nvidia_rtx_4060_ti_16gb": 3054.5, "nvidia_rtx_4070_ti_super": 7127.3, "nvidia_rtx_4080_super": 7806.1, "nvidia_rtx_4090": 10690.9, "nvidia_rtx_5080": 10181.8, "nvidia_rtx_5090": 19006.1}, "model_id": "microsoft/xclip-base-patch32", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Microsoft", "scores_as_of": null}, {"benchmark_scores": {"live_code_bench": 81.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 204800, "cost_input": 0.3, "display_name": "MiniMax-M2.1", "fits": {}, "model_id": "minimax/minimax-m2-1", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "MiniMax", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 204800, "cost_input": 0.6, "display_name": "MiniMax-M2.5-highspeed", "fits": {}, "model_id": "minimax/minimax-m2-5-highspeed", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "MiniMax", "scores_as_of": null}, {"benchmark_scores": {"live_code_bench": 82.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 204800, "cost_input": 0.3, "display_name": "MiniMax-M2.5", "fits": {}, "model_id": "minimax/minimax-m2-5", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "MiniMax", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 204800, "cost_input": 0.6, "display_name": "MiniMax-M2.7-highspeed", "fits": {}, "model_id": "minimax/minimax-m2-7-highspeed", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "MiniMax", "scores_as_of": null}, {"benchmark_scores": {"live_code_bench": 82.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 204800, "cost_input": 0.3, "display_name": "MiniMax-M2.7", "fits": {}, "model_id": "minimax/minimax-m2-7", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "MiniMax", "scores_as_of": "2026-04"}, {"benchmark_scores": {"live_code_bench": 82.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 196608, "cost_input": 0.3, "display_name": "MiniMax-M2", "fits": {}, "model_id": "minimax/minimax-m2", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "MiniMax", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 60.5, "arena_elo_coding": 1290.0, "arena_elo_math": 1240.0, "arena_elo_overall": 1260.0, "gpqa_diamond": 42.5, "humaneval": 88.5, "ifeval": 74.0, "live_code_bench": 31.4, "math_500": 65.2, "mmlu_pro": 58.5, "multipl_e": 75.7, "multipl_e_cpp": 75.2, "multipl_e_csharp": 73.5, "multipl_e_go": 72.5, "multipl_e_java": 79.5, "multipl_e_javascript": 78.2, "multipl_e_kotlin": 61.5, "multipl_e_lua": 48.5, "multipl_e_php": 71.2, "multipl_e_python": 81.5, "multipl_e_ruby": 55.2, "multipl_e_rust": 65.8, "multipl_e_scala": 52.1, "multipl_e_swift": 56.8, "multipl_e_typescript": 77.1, "swe_bench_verified": 35.2}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.3, "display_name": "Codestral (latest)", "fits": {"amd_rx_7900_xt": 50.9, "amd_rx_7900_xtx": 61.1, "amd_rx_9070_xt": 41.0, "apple_m1_max": 25.5, "apple_m2_max": 25.5, "apple_m2_ultra": 50.9, "apple_m3_max": 25.5, "apple_m3_ultra": 52.1, "apple_m4": 7.6, "apple_m4_max": 34.7, "apple_m4_pro": 17.4, "nvidia_rtx_3090": 59.6, "nvidia_rtx_4060_ti_16gb": 18.3, "nvidia_rtx_4070_ti_super": 42.8, "nvidia_rtx_4080_super": 46.8, "nvidia_rtx_4090": 64.1, "nvidia_rtx_5080": 61.1, "nvidia_rtx_5090": 114.0}, "model_id": "mistral/codestral-latest", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "Devstral 2 123B Instruct 2512", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.1, "apple_m3_max": 4.6, "apple_m3_ultra": 9.3, "apple_m4_max": 6.2}, "model_id": "mistral/devstral-2-123b-instruct-2512", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.4, "display_name": "Devstral 2", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/devstral-2512", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {"swe_bench_verified": 61.6}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.4, "display_name": "Devstral Medium", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/devstral-medium-2507", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"swe_bench_verified": 61.6}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.4, "display_name": "Devstral 2 (latest)", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/devstral-medium-latest", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"swe_bench_verified": 53.6}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "Devstral Small 2 24B Instruct 2512", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/devstral-small-2-24b-instruct-2512", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"swe_bench_verified": 53.6}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.1, "display_name": "Devstral Small 2505", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/devstral-small-2505", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"swe_bench_verified": 53.6}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.1, "display_name": "Devstral Small", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/devstral-small-2507", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.0, "display_name": "Devstral Small 2", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/labs-devstral-small-2512", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1300.0, "arena_elo_hard_prompts": 1331.2, "arena_elo_math": 1320.0, "arena_elo_overall": 1310.0, "arena_elo_style_control": 1303.0, "gpqa_diamond": 60.2, "humaneval": 86.5, "ifeval": 85.0, "live_code_bench": 50.3, "math_500": 85.5, "mmlu_pro": 74.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 2.0, "display_name": "Magistral Medium (latest)", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.1, "apple_m3_max": 4.6, "apple_m3_ultra": 9.3, "apple_m4_max": 6.2}, "model_id": "mistral/magistral-medium-latest", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1280.0, "arena_elo_overall": 1270.0, "gpqa_diamond": 50.5, "humaneval": 78.5, "ifeval": 78.0, "live_code_bench": 51.3, "math_500": 78.2, "mmlu_pro": 64.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Magistral Small 2506", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/magistral-small-2506", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1280.0, "arena_elo_overall": 1270.0, "gpqa_diamond": 50.5, "humaneval": 78.5, "ifeval": 78.0, "live_code_bench": 51.3, "math_500": 78.2, "mmlu_pro": 64.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.5, "display_name": "Magistral Small", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/magistral-small", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 256000, "cost_input": 0.3, "display_name": "Mamba Codestral 7B v0.1", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mamba-codestral-7b-v0-1", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Ministral 3 14B Instruct 2512", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "mistral/ministral-3-14b-instruct-2512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Ministral 3 3B Base 2512", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "mistral/ministral-3-3b-base-2512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Ministral 3 3B Instruct 2512 BF16", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "mistral/ministral-3-3b-instruct-2512-bf16", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Ministral 3 3B Instruct 2512", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "mistral/ministral-3-3b-instruct-2512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Ministral 3 8B Instruct 2512 BF16", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "mistral/ministral-3-8b-instruct-2512-bf16", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Ministral 3 8B Instruct 2512", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "mistral/ministral-3-8b-instruct-2512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 31.9, "gpqa_diamond": 25.2, "ifeval": 13.6, "math_500": 0.8, "mmlu_pro": 10.9, "musr": 33.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.04, "display_name": "Ministral 3B (latest)", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "mistral/ministral-3b-latest", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 47.6, "gpqa_diamond": 28.4, "ifeval": 59.0, "math_500": 19.6, "mmlu_pro": 32.9, "musr": 41.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Ministral 8B Instruct 2410", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "mistral/ministral-8b-instruct-2410", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.1, "display_name": "Ministral 8B (latest)", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "mistral/ministral-8b-latest", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 33.5, "gpqa_diamond": 25.0, "gsm8k": 14.3, "ifeval": 44.9, "math_500": 2.3, "mmlu_pro": 24.1, "musr": 38.5, "winogrande": 73.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Mistral 7B Instruct v0.1", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mistral-7b-instruct-v0-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 63.1, "bbh": 44.6, "gpqa_diamond": 27.6, "gsm8k": 40.0, "hellaswag": 84.9, "ifeval": 55.0, "math_500": 3.0, "mmlu_abstract_algebra": 32.0, "mmlu_anatomy": 57.0, "mmlu_astronomy": 62.5, "mmlu_business_ethics": 60.0, "mmlu_clinical_knowledge": 67.2, "mmlu_college_biology": 69.4, "mmlu_college_chemistry": 40.0, "mmlu_college_computer_science": 53.0, "mmlu_college_mathematics": 39.0, "mmlu_college_medicine": 59.0, "mmlu_college_physics": 42.2, "mmlu_computer_security": 69.0, "mmlu_conceptual_physics": 53.2, "mmlu_econometrics": 40.4, "mmlu_electrical_engineering": 61.4, "mmlu_elementary_mathematics": 36.8, "mmlu_formal_logic": 42.1, "mmlu_global_facts": 35.0, "mmlu_high_school_biology": 63.2, "mmlu_high_school_chemistry": 51.2, "mmlu_high_school_computer_science": 65.0, "mmlu_high_school_european_history": 73.9, "mmlu_high_school_geography": 76.3, "mmlu_high_school_government_and_politics": 85.5, "mmlu_high_school_macroeconomics": 55.6, "mmlu_high_school_mathematics": 30.0, "mmlu_high_school_microeconomics": 65.5, "mmlu_high_school_physics": 35.8, "mmlu_high_school_psychology": 79.1, "mmlu_high_school_statistics": 44.4, "mmlu_high_school_us_history": 76.5, "mmlu_high_school_world_history": 75.5, "mmlu_human_aging": 61.9, "mmlu_human_sexuality": 74.0, "mmlu_international_law": 81.0, "mmlu_jurisprudence": 74.1, "mmlu_logical_fallacies": 73.0, "mmlu_machine_learning": 44.6, "mmlu_management": 75.7, "mmlu_marketing": 86.3, "mmlu_medical_genetics": 67.0, "mmlu_miscellaneous": 78.0, "mmlu_moral_disputes": 69.7, "mmlu_moral_scenarios": 31.2, "mmlu_nutrition": 68.6, "mmlu_philosophy": 70.4, "mmlu_prehistory": 70.1, "mmlu_pro": 27.2, "mmlu_professional_accounting": 45.0, "mmlu_professional_law": 43.5, "mmlu_professional_medicine": 61.8, "mmlu_professional_psychology": 63.1, "mmlu_public_relations": 70.9, "mmlu_security_studies": 70.6, "mmlu_sociology": 73.1, "mmlu_us_foreign_policy": 81.0, "mmlu_virology": 49.4, "mmlu_world_religions": 83.6, "musr": 39.7, "truthfulqa": 68.3, "winogrande": 77.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Mistral 7B Instruct v0.2", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mistral-7b-instruct-v0-2", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 47.2, "gpqa_diamond": 27.9, "ifeval": 54.7, "math_500": 3.9, "mmlu_pro": 30.8, "musr": 37.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Mistral 7B Instruct v0.3", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mistral-7b-instruct-v0-3", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 44.2, "gpqa_diamond": 29.2, "gsm8k": 37.1, "ifeval": 23.9, "math_500": 2.9, "mmlu_pro": 30.1, "musr": 41.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Mistral 7B v0.1", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mistral-7b-v0-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 60.5, "bbh": 45.2, "gpqa_diamond": 29.2, "gsm8k": 34.5, "hellaswag": 83.0, "ifeval": 22.7, "math_500": 3.0, "mmlu_abstract_algebra": 30.0, "mmlu_anatomy": 60.7, "mmlu_astronomy": 65.1, "mmlu_business_ethics": 61.0, "mmlu_clinical_knowledge": 68.3, "mmlu_college_biology": 69.4, "mmlu_college_chemistry": 49.0, "mmlu_college_computer_science": 52.0, "mmlu_college_mathematics": 37.0, "mmlu_college_medicine": 63.0, "mmlu_college_physics": 31.4, "mmlu_computer_security": 73.0, "mmlu_conceptual_physics": 58.3, "mmlu_econometrics": 44.7, "mmlu_electrical_engineering": 55.9, "mmlu_elementary_mathematics": 40.2, "mmlu_formal_logic": 39.7, "mmlu_global_facts": 40.0, "mmlu_high_school_biology": 76.1, "mmlu_high_school_chemistry": 54.2, "mmlu_high_school_computer_science": 65.0, "mmlu_high_school_european_history": 76.4, "mmlu_high_school_geography": 77.3, "mmlu_high_school_government_and_politics": 87.6, "mmlu_high_school_macroeconomics": 60.5, "mmlu_high_school_mathematics": 35.9, "mmlu_high_school_microeconomics": 63.9, "mmlu_high_school_physics": 35.8, "mmlu_high_school_psychology": 80.0, "mmlu_high_school_statistics": 51.9, "mmlu_high_school_us_history": 81.4, "mmlu_high_school_world_history": 78.1, "mmlu_human_aging": 70.0, "mmlu_human_sexuality": 77.1, "mmlu_international_law": 81.0, "mmlu_jurisprudence": 74.1, "mmlu_logical_fallacies": 78.5, "mmlu_machine_learning": 53.6, "mmlu_management": 81.6, "mmlu_marketing": 88.0, "mmlu_medical_genetics": 70.0, "mmlu_miscellaneous": 79.8, "mmlu_moral_disputes": 70.2, "mmlu_moral_scenarios": 39.8, "mmlu_nutrition": 74.2, "mmlu_philosophy": 72.7, "mmlu_prehistory": 68.8, "mmlu_pro": 29.5, "mmlu_professional_accounting": 51.1, "mmlu_professional_law": 46.2, "mmlu_professional_medicine": 68.8, "mmlu_professional_psychology": 66.2, "mmlu_public_relations": 64.5, "mmlu_security_studies": 72.7, "mmlu_sociology": 82.6, "mmlu_us_foreign_policy": 88.0, "mmlu_virology": 53.6, "mmlu_world_religions": 81.9, "musr": 40.3, "truthfulqa": 41.8, "winogrande": 78.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Mistral 7B v0.3", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mistral-7b-v0-3", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {"beir": 49.2, "miracl": 45.0, "mteb_classification": 66.1, "mteb_clustering": 44.5, "mteb_overall": 60.2, "mteb_pair_classification": 82.0, "mteb_reranking": 54.0, "mteb_retrieval": 55.8, "mteb_sts": 79.0, "mteb_summarization": 28.0}, "capability_tiers": {}, "context_window": 8000, "cost_input": 0.1, "display_name": "Mistral Embed", "fits": {}, "model_id": "mistral/mistral-embed", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 52.8, "alpaca_eval": 38.5, "arena_elo_coding": 1270.0, "arena_elo_hard_prompts": 1312.6, "arena_elo_math": 1260.0, "arena_elo_overall": 1280.0, "arena_elo_style_control": 1304.7, "artificial_analysis_quality_index": 78.0, "artificial_analysis_speed_index": 60.0, "bbh": 58.9, "bbq": 77.5, "gpqa_diamond": 55.5, "helm_safety": 82.8, "humaneval": 84.2, "ifeval": 82.0, "live_code_bench": 34.4, "math_500": 78.5, "medqa": 81.5, "mmlu_astronomy": 73.5, "mmlu_biology": 81.5, "mmlu_business_ethics": 74.8, "mmlu_chemistry": 74.8, "mmlu_clinical_knowledge": 80.5, "mmlu_computer_science": 82.2, "mmlu_jurisprudence": 71.2, "mmlu_physics": 78.2, "mmlu_pro": 69.8, "mmlu_professional_accounting": 65.8, "mmlu_professional_law": 70.5, "mt_bench": 8.5, "multipl_e": 75.8, "multipl_e_cpp": 75.5, "multipl_e_csharp": 78.2, "multipl_e_go": 72.1, "multipl_e_java": 80.2, "multipl_e_javascript": 78.8, "multipl_e_julia": 58.8, "multipl_e_kotlin": 70.2, "multipl_e_lua": 52.8, "multipl_e_perl": 48.5, "multipl_e_php": 75.5, "multipl_e_python": 82.8, "multipl_e_r": 55.2, "multipl_e_ruby": 62.5, "multipl_e_rust": 63.5, "multipl_e_scala": 57.8, "multipl_e_swift": 63.5, "multipl_e_typescript": 77.5, "musr": 26.3, "swe_bench_verified": 32.5, "toxigen": 86.2, "wildbench": 65.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 2.0, "display_name": "Mistral Large 2.1", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.1, "apple_m3_max": 4.6, "apple_m3_ultra": 9.3, "apple_m4_max": 6.2}, "model_id": "mistral/mistral-large-2411", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 52.8, "alpaca_eval": 38.5, "arena_elo_coding": 1270.0, "arena_elo_hard_prompts": 1431.1, "arena_elo_math": 1260.0, "arena_elo_overall": 1280.0, "arena_elo_style_control": 1415.0, "artificial_analysis_quality_index": 78.0, "artificial_analysis_speed_index": 60.0, "bbh": 58.9, "bbq": 77.5, "chartqa": 74.5, "docvqa": 83.5, "gpqa_diamond": 55.5, "helm_safety": 82.8, "humaneval": 84.2, "ifeval": 82.0, "live_code_bench": 34.4, "math_500": 78.5, "mathvista": 50.2, "mmlu_astronomy": 73.5, "mmlu_biology": 81.5, "mmlu_business_ethics": 74.8, "mmlu_chemistry": 74.8, "mmlu_clinical_knowledge": 80.5, "mmlu_computer_science": 82.2, "mmlu_jurisprudence": 71.2, "mmlu_physics": 78.2, "mmlu_pro": 69.8, "mmlu_professional_accounting": 65.8, "mmlu_professional_law": 70.5, "mmmu": 56.8, "mt_bench": 8.5, "multipl_e": 75.8, "multipl_e_cpp": 75.5, "multipl_e_csharp": 78.2, "multipl_e_go": 72.1, "multipl_e_java": 80.2, "multipl_e_javascript": 78.8, "multipl_e_julia": 58.8, "multipl_e_kotlin": 70.2, "multipl_e_lua": 52.8, "multipl_e_perl": 48.5, "multipl_e_php": 75.5, "multipl_e_python": 82.8, "multipl_e_r": 55.2, "multipl_e_ruby": 62.5, "multipl_e_rust": 63.5, "multipl_e_scala": 57.8, "multipl_e_swift": 63.5, "multipl_e_typescript": 77.5, "musr": 26.3, "swe_bench_verified": 32.5, "toxigen": 86.2, "wildbench": 65.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.5, "display_name": "Mistral Large 3", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.1, "apple_m3_max": 4.6, "apple_m3_ultra": 9.3, "apple_m4_max": 6.2}, "model_id": "mistral/mistral-large-2512", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 52.8, "alpaca_eval": 38.5, "arena_elo_coding": 1270.0, "arena_elo_math": 1260.0, "arena_elo_overall": 1280.0, "artificial_analysis_quality_index": 78.0, "artificial_analysis_speed_index": 60.0, "bbh": 58.9, "bbq": 77.5, "chartqa": 74.5, "docvqa": 83.5, "gpqa_diamond": 55.5, "gsm8k": 93.0, "helm_safety": 82.8, "humaneval": 84.2, "ifeval": 82.0, "live_code_bench": 34.4, "math_500": 78.5, "mathvista": 50.2, "mmlu_astronomy": 73.5, "mmlu_biology": 81.5, "mmlu_business_ethics": 74.8, "mmlu_chemistry": 74.8, "mmlu_clinical_knowledge": 80.5, "mmlu_computer_science": 82.2, "mmlu_jurisprudence": 71.2, "mmlu_physics": 78.2, "mmlu_pro": 69.8, "mmlu_professional_accounting": 65.8, "mmlu_professional_law": 70.5, "mmmu": 56.8, "mt_bench": 8.5, "multipl_e": 75.8, "multipl_e_cpp": 75.5, "multipl_e_csharp": 78.2, "multipl_e_go": 72.1, "multipl_e_java": 80.2, "multipl_e_javascript": 78.8, "multipl_e_julia": 58.8, "multipl_e_kotlin": 70.2, "multipl_e_lua": 52.8, "multipl_e_perl": 48.5, "multipl_e_php": 75.5, "multipl_e_python": 82.8, "multipl_e_r": 55.2, "multipl_e_ruby": 62.5, "multipl_e_rust": 63.5, "multipl_e_scala": 57.8, "multipl_e_swift": 63.5, "multipl_e_typescript": 77.5, "musr": 26.3, "swe_bench_verified": 32.5, "toxigen": 86.2, "wildbench": 65.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.5, "display_name": "Mistral Large (latest)", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.1, "apple_m3_max": 4.6, "apple_m3_ultra": 9.3, "apple_m4_max": 6.2}, "model_id": "mistral/mistral-large-latest", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1240.0, "arena_elo_hard_prompts": 1401.1, "arena_elo_math": 1230.0, "arena_elo_overall": 1250.0, "arena_elo_style_control": 1386.2, "arena_elo_vision": 1156.6, "legalbench": 89.4, "medqa": 79.1, "multipl_e_csharp": 72.5, "multipl_e_julia": 52.5, "multipl_e_kotlin": 63.5, "multipl_e_lua": 47.2, "multipl_e_perl": 42.5, "multipl_e_php": 70.2, "multipl_e_r": 48.5, "multipl_e_ruby": 56.5, "multipl_e_scala": 51.2, "multipl_e_swift": 56.8}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.4, "display_name": "Mistral Medium 3", "fits": {}, "model_id": "mistral/mistral-medium-2505", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1240.0, "arena_elo_hard_prompts": 1429.3, "arena_elo_math": 1230.0, "arena_elo_overall": 1250.0, "arena_elo_style_control": 1410.3, "arena_elo_vision": 1157.3, "legalbench": 89.4, "medqa": 79.1, "multipl_e_csharp": 72.5, "multipl_e_julia": 52.5, "multipl_e_kotlin": 63.5, "multipl_e_lua": 47.2, "multipl_e_perl": 42.5, "multipl_e_php": 70.2, "multipl_e_r": 48.5, "multipl_e_ruby": 56.5, "multipl_e_scala": 51.2, "multipl_e_swift": 56.8}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.4, "display_name": "Mistral Medium 3.1", "fits": {}, "model_id": "mistral/mistral-medium-2508", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"alpaca_eval": 28.6, "arena_elo_coding": 1240.0, "arena_elo_math": 1230.0, "arena_elo_overall": 1250.0, "legalbench": 89.4, "medqa": 79.1, "multipl_e_csharp": 72.5, "multipl_e_julia": 52.5, "multipl_e_kotlin": 63.5, "multipl_e_lua": 47.2, "multipl_e_perl": 42.5, "multipl_e_php": 70.2, "multipl_e_r": 48.5, "multipl_e_ruby": 56.5, "multipl_e_scala": 51.2, "multipl_e_swift": 56.8}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.4, "display_name": "Mistral Medium (latest)", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.1, "apple_m3_max": 4.6, "apple_m3_ultra": 9.3, "apple_m4_max": 6.2}, "model_id": "mistral/mistral-medium-latest", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"artificial_analysis_quality_index": 65.0, "artificial_analysis_speed_index": 85.0, "bbh": 44.2, "gpqa_diamond": 33.5, "ifeval": 75.1, "math_500": 55.3, "mmlu_pro": 47.8, "musr": 16.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.15, "display_name": "Mistral Nemo Base 2407", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "mistral/mistral-nemo-base-2407", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"artificial_analysis_quality_index": 65.0, "artificial_analysis_speed_index": 85.0, "bbh": 44.2, "gpqa_diamond": 33.5, "ifeval": 75.1, "math_500": 55.3, "mmlu_pro": 47.8, "musr": 16.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.15, "display_name": "Mistral Nemo Instruct 2407", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "mistral/mistral-nemo-instruct-2407", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"artificial_analysis_quality_index": 65.0, "artificial_analysis_speed_index": 85.0, "bbh": 44.2, "gpqa_diamond": 33.5, "ifeval": 75.1, "math_500": 55.3, "mmlu_pro": 47.8, "musr": 16.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.15, "display_name": "Mistral Nemo", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "mistral/mistral-nemo", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_hard_prompts": 1284.3, "arena_elo_math": 1200.0, "arena_elo_overall": 1220.0, "arena_elo_style_control": 1273.5, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 82.0, "multipl_e_csharp": 65.2, "multipl_e_julia": 44.2, "multipl_e_kotlin": 55.5, "multipl_e_lua": 40.5, "multipl_e_perl": 35.2, "multipl_e_php": 62.5, "multipl_e_r": 40.5, "multipl_e_ruby": 48.5, "multipl_e_scala": 43.2, "multipl_e_swift": 48.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.2, "display_name": "Mistral Small 24B Instruct 2501", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/mistral-small-24b-instruct-2501", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_hard_prompts": 1374.3, "arena_elo_math": 1200.0, "arena_elo_overall": 1220.0, "arena_elo_style_control": 1357.0, "arena_elo_vision": 1140.1, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 82.0, "multipl_e_csharp": 65.2, "multipl_e_julia": 44.2, "multipl_e_kotlin": 55.5, "multipl_e_lua": 40.5, "multipl_e_perl": 35.2, "multipl_e_php": 62.5, "multipl_e_r": 40.5, "multipl_e_ruby": 48.5, "multipl_e_scala": 43.2, "multipl_e_swift": 48.8}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.1, "display_name": "Mistral Small 3.2", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/mistral-small-2506", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1200.0, "arena_elo_overall": 1220.0, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 82.0, "multipl_e_csharp": 65.2, "multipl_e_julia": 44.2, "multipl_e_kotlin": 55.5, "multipl_e_lua": 40.5, "multipl_e_perl": 35.2, "multipl_e_php": 62.5, "multipl_e_r": 40.5, "multipl_e_ruby": 48.5, "multipl_e_scala": 43.2, "multipl_e_swift": 48.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.15, "display_name": "Mistral Small 4", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/mistral-small-2603", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_hard_prompts": 1318.7, "arena_elo_math": 1200.0, "arena_elo_overall": 1220.0, "arena_elo_style_control": 1302.7, "arena_elo_vision": 1128.1, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 82.0, "multipl_e_csharp": 65.2, "multipl_e_julia": 44.2, "multipl_e_kotlin": 55.5, "multipl_e_lua": 40.5, "multipl_e_perl": 35.2, "multipl_e_php": 62.5, "multipl_e_r": 40.5, "multipl_e_ruby": 48.5, "multipl_e_scala": 43.2, "multipl_e_swift": 48.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "Mistral Small 3.1 24B Instruct 2503", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/mistral-small-3-1-24b-instruct-2503", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1200.0, "arena_elo_overall": 1220.0, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 82.0, "multipl_e_csharp": 65.2, "multipl_e_julia": 44.2, "multipl_e_kotlin": 55.5, "multipl_e_lua": 40.5, "multipl_e_perl": 35.2, "multipl_e_php": 62.5, "multipl_e_r": 40.5, "multipl_e_ruby": 48.5, "multipl_e_scala": 43.2, "multipl_e_swift": 48.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "Mistral Small 3.2 24B Instruct 2506", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/mistral-small-3-2-24b-instruct-2506", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1200.0, "arena_elo_overall": 1220.0, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 82.0, "multipl_e_csharp": 65.2, "multipl_e_julia": 44.2, "multipl_e_kotlin": 55.5, "multipl_e_lua": 40.5, "multipl_e_perl": 35.2, "multipl_e_php": 62.5, "multipl_e_r": 40.5, "multipl_e_ruby": 48.5, "multipl_e_scala": 43.2, "multipl_e_swift": 48.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "Mistral Small 4 119B 2603", "fits": {"apple_m2_max": 4.7, "apple_m2_ultra": 9.4, "apple_m3_max": 4.7, "apple_m3_ultra": 9.6, "apple_m4_max": 6.4}, "model_id": "mistral/mistral-small-4-119b-2603", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1200.0, "arena_elo_overall": 1220.0, "artificial_analysis_quality_index": 68.0, "artificial_analysis_speed_index": 82.0, "multipl_e_csharp": 65.2, "multipl_e_julia": 44.2, "multipl_e_kotlin": 55.5, "multipl_e_lua": 40.5, "multipl_e_perl": 35.2, "multipl_e_php": 62.5, "multipl_e_r": 40.5, "multipl_e_ruby": 48.5, "multipl_e_scala": 43.2, "multipl_e_swift": 48.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.15, "display_name": "Mistral Small (latest)", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/mistral-small-latest", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 72.7, "bbh": 48.5, "gpqa_diamond": 36.1, "gsm8k": 88.0, "hellaswag": 89.1, "ifeval": 77.8, "math_500": 56.2, "mmlu_abstract_algebra": 54.0, "mmlu_anatomy": 74.8, "mmlu_astronomy": 86.2, "mmlu_business_ethics": 77.0, "mmlu_clinical_knowledge": 82.3, "mmlu_college_biology": 89.6, "mmlu_college_chemistry": 56.0, "mmlu_college_computer_science": 70.0, "mmlu_college_mathematics": 49.0, "mmlu_college_medicine": 76.3, "mmlu_college_physics": 57.8, "mmlu_computer_security": 81.0, "mmlu_conceptual_physics": 79.1, "mmlu_econometrics": 63.2, "mmlu_electrical_engineering": 75.9, "mmlu_elementary_mathematics": 62.4, "mmlu_formal_logic": 59.5, "mmlu_global_facts": 54.0, "mmlu_high_school_biology": 90.0, "mmlu_high_school_chemistry": 69.5, "mmlu_high_school_computer_science": 86.0, "mmlu_high_school_european_history": 85.5, "mmlu_high_school_geography": 89.4, "mmlu_high_school_government_and_politics": 96.9, "mmlu_high_school_macroeconomics": 81.3, "mmlu_high_school_mathematics": 50.7, "mmlu_high_school_microeconomics": 87.8, "mmlu_high_school_physics": 50.3, "mmlu_high_school_psychology": 92.7, "mmlu_high_school_statistics": 69.4, "mmlu_high_school_us_history": 89.7, "mmlu_high_school_world_history": 91.6, "mmlu_human_aging": 80.3, "mmlu_human_sexuality": 88.5, "mmlu_international_law": 90.1, "mmlu_jurisprudence": 86.1, "mmlu_logical_fallacies": 87.1, "mmlu_machine_learning": 61.6, "mmlu_management": 87.4, "mmlu_marketing": 92.7, "mmlu_medical_genetics": 84.0, "mmlu_miscellaneous": 89.9, "mmlu_moral_disputes": 84.7, "mmlu_moral_scenarios": 66.0, "mmlu_nutrition": 87.9, "mmlu_philosophy": 82.6, "mmlu_prehistory": 87.0, "mmlu_pro": 52.3, "mmlu_professional_accounting": 66.3, "mmlu_professional_law": 60.3, "mmlu_professional_medicine": 88.6, "mmlu_professional_psychology": 84.0, "mmlu_public_relations": 77.3, "mmlu_security_studies": 84.5, "mmlu_sociology": 91.5, "mmlu_us_foreign_policy": 96.0, "mmlu_virology": 59.0, "mmlu_world_religions": 90.6, "musr": 18.5, "truthfulqa": 68.1, "winogrande": 85.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 65536, "cost_input": 0.65, "display_name": "Mixtral 8x22B Instruct v0.1", "fits": {"amd_rx_7900_xt": 50.9, "amd_rx_7900_xtx": 61.1, "amd_rx_9070_xt": 41.0, "apple_m1_max": 25.5, "apple_m2_max": 25.5, "apple_m2_ultra": 50.9, "apple_m3_max": 25.5, "apple_m3_ultra": 52.1, "apple_m4": 7.6, "apple_m4_max": 34.7, "apple_m4_pro": 17.4, "nvidia_rtx_3090": 59.6, "nvidia_rtx_4060_ti_16gb": 18.3, "nvidia_rtx_4070_ti_super": 42.8, "nvidia_rtx_4080_super": 46.8, "nvidia_rtx_4090": 64.1, "nvidia_rtx_5080": 61.1, "nvidia_rtx_5090": 114.0}, "model_id": "mistral/mixtral-8x22b-instruct-v0-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 70.1, "bbh": 36.8, "gpqa_diamond": 27.2, "gsm8k": 61.1, "hellaswag": 87.6, "ifeval": 69.5, "math_500": 40.1, "mmlu_abstract_algebra": 42.0, "mmlu_anatomy": 66.7, "mmlu_astronomy": 78.9, "mmlu_business_ethics": 73.0, "mmlu_clinical_knowledge": 77.7, "mmlu_college_biology": 82.6, "mmlu_college_chemistry": 50.0, "mmlu_college_computer_science": 66.0, "mmlu_college_mathematics": 46.0, "mmlu_college_medicine": 75.7, "mmlu_college_physics": 43.1, "mmlu_computer_security": 81.0, "mmlu_conceptual_physics": 66.8, "mmlu_econometrics": 61.4, "mmlu_electrical_engineering": 64.8, "mmlu_elementary_mathematics": 47.9, "mmlu_formal_logic": 52.4, "mmlu_global_facts": 42.0, "mmlu_high_school_biology": 85.2, "mmlu_high_school_chemistry": 62.1, "mmlu_high_school_computer_science": 78.0, "mmlu_high_school_european_history": 80.0, "mmlu_high_school_geography": 86.9, "mmlu_high_school_government_and_politics": 95.9, "mmlu_high_school_macroeconomics": 69.7, "mmlu_high_school_mathematics": 38.9, "mmlu_high_school_microeconomics": 80.3, "mmlu_high_school_physics": 47.0, "mmlu_high_school_psychology": 88.4, "mmlu_high_school_statistics": 59.7, "mmlu_high_school_us_history": 85.3, "mmlu_high_school_world_history": 84.8, "mmlu_human_aging": 75.8, "mmlu_human_sexuality": 80.9, "mmlu_international_law": 87.6, "mmlu_jurisprudence": 84.3, "mmlu_logical_fallacies": 81.6, "mmlu_machine_learning": 57.1, "mmlu_management": 84.5, "mmlu_marketing": 92.3, "mmlu_medical_genetics": 77.0, "mmlu_miscellaneous": 88.0, "mmlu_moral_disputes": 78.0, "mmlu_moral_scenarios": 46.0, "mmlu_nutrition": 82.0, "mmlu_philosophy": 79.7, "mmlu_prehistory": 83.3, "mmlu_pro": 40.5, "mmlu_professional_accounting": 55.3, "mmlu_professional_law": 54.4, "mmlu_professional_medicine": 79.4, "mmlu_professional_psychology": 76.5, "mmlu_public_relations": 70.9, "mmlu_security_studies": 77.1, "mmlu_sociology": 89.1, "mmlu_us_foreign_policy": 90.0, "mmlu_virology": 50.6, "mmlu_world_religions": 87.7, "musr": 11.8, "truthfulqa": 65.0, "winogrande": 81.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.24, "display_name": "Mixtral 8x7B Instruct v0.1", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mixtral-8x7b-instruct-v0-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 66.4, "bbh": 36.8, "gpqa_diamond": 27.2, "gsm8k": 57.6, "hellaswag": 86.5, "ifeval": 69.5, "math_500": 40.1, "mmlu_abstract_algebra": 34.0, "mmlu_anatomy": 71.9, "mmlu_astronomy": 82.9, "mmlu_business_ethics": 76.0, "mmlu_clinical_knowledge": 78.5, "mmlu_college_biology": 86.8, "mmlu_college_chemistry": 54.0, "mmlu_college_computer_science": 63.0, "mmlu_college_mathematics": 46.0, "mmlu_college_medicine": 69.9, "mmlu_college_physics": 46.1, "mmlu_computer_security": 81.0, "mmlu_conceptual_physics": 68.1, "mmlu_econometrics": 64.9, "mmlu_electrical_engineering": 69.0, "mmlu_elementary_mathematics": 48.1, "mmlu_formal_logic": 56.3, "mmlu_global_facts": 51.0, "mmlu_high_school_biology": 84.2, "mmlu_high_school_chemistry": 63.5, "mmlu_high_school_computer_science": 72.0, "mmlu_high_school_european_history": 81.8, "mmlu_high_school_geography": 86.4, "mmlu_high_school_government_and_politics": 93.8, "mmlu_high_school_macroeconomics": 70.5, "mmlu_high_school_mathematics": 38.5, "mmlu_high_school_microeconomics": 78.6, "mmlu_high_school_physics": 49.0, "mmlu_high_school_psychology": 88.1, "mmlu_high_school_statistics": 64.8, "mmlu_high_school_us_history": 84.8, "mmlu_high_school_world_history": 89.0, "mmlu_human_aging": 78.0, "mmlu_human_sexuality": 80.9, "mmlu_international_law": 87.6, "mmlu_jurisprudence": 83.3, "mmlu_logical_fallacies": 77.3, "mmlu_machine_learning": 53.6, "mmlu_management": 88.3, "mmlu_marketing": 91.9, "mmlu_medical_genetics": 78.0, "mmlu_miscellaneous": 87.5, "mmlu_moral_disputes": 79.8, "mmlu_moral_scenarios": 40.1, "mmlu_nutrition": 82.4, "mmlu_philosophy": 78.5, "mmlu_prehistory": 84.0, "mmlu_pro": 40.5, "mmlu_professional_accounting": 51.8, "mmlu_professional_law": 53.2, "mmlu_professional_medicine": 81.2, "mmlu_professional_psychology": 78.4, "mmlu_public_relations": 70.0, "mmlu_security_studies": 78.8, "mmlu_sociology": 89.1, "mmlu_us_foreign_policy": 92.0, "mmlu_virology": 51.2, "mmlu_world_religions": 87.7, "musr": 11.8, "truthfulqa": 46.8, "winogrande": 81.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.24, "display_name": "Mixtral 8x7B v0.1", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/mixtral-8x7b-v0-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8000, "cost_input": 0.25, "display_name": "Mistral 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "mistral/open-mistral-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 48.5, "gpqa_diamond": 36.1, "gsm8k": 88.0, "ifeval": 77.8, "math_500": 56.2, "mmlu_pro": 52.3, "musr": 18.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 64000, "cost_input": 2.0, "display_name": "Mixtral 8x22B", "fits": {"apple_m2_max": 4.0, "apple_m2_ultra": 7.9, "apple_m3_max": 4.0, "apple_m3_ultra": 8.1, "apple_m4_max": 5.4}, "model_id": "mistral/open-mixtral-8x22b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 36.8, "gpqa_diamond": 27.2, "ifeval": 69.5, "math_500": 40.1, "mmlu_pro": 40.5, "musr": 11.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32000, "cost_input": 0.7, "display_name": "Mixtral 8x7B", "fits": {"apple_m1_max": 12.0, "apple_m2_max": 12.0, "apple_m2_ultra": 24.0, "apple_m3_max": 12.0, "apple_m3_ultra": 24.6, "apple_m4": 3.6, "apple_m4_max": 16.4, "apple_m4_pro": 8.2, "nvidia_rtx_5090": 53.7}, "model_id": "mistral/open-mixtral-8x7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 79.5, "bbq": 78.5, "chartqa": 66.5, "docvqa": 78.8, "helm_safety": 83.5, "mathvista": 43.8, "mmmu": 50.2, "ocrbench": 68.9, "toxigen": 86.8}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.15, "display_name": "Pixtral 12B", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "mistral/pixtral-12b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 93.8, "bbq": 82.2, "chartqa": 76.2, "docvqa": 85.5, "helm_safety": 86.8, "mathvista": 52.5, "medqa": 78.3, "mmmu": 58.8, "ocrbench": 78.5, "toxigen": 90.5}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 2.0, "display_name": "Pixtral Large (latest)", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.1, "apple_m3_max": 4.6, "apple_m3_ultra": 9.3, "apple_m4_max": 6.2}, "model_id": "mistral/pixtral-large-latest", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Mistral AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Voxtral Mini 3B 2507", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "mistral/voxtral-mini-3b-2507", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Voxtral Mini 4B Realtime 2602", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "mistral/voxtral-mini-4b-realtime-2602", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Voxtral Small 24B 2507", "fits": {"amd_rx_7900_xt": 46.7, "amd_rx_7900_xtx": 56.0, "amd_rx_9070_xt": 37.6, "apple_m1_max": 23.3, "apple_m2_max": 23.3, "apple_m2_ultra": 46.7, "apple_m3_max": 23.3, "apple_m3_ultra": 47.8, "apple_m4": 7.0, "apple_m4_max": 31.8, "apple_m4_pro": 15.9, "nvidia_rtx_3090": 54.6, "nvidia_rtx_4060_ti_16gb": 16.8, "nvidia_rtx_4070_ti_super": 39.2, "nvidia_rtx_4080_super": 42.9, "nvidia_rtx_4090": 58.8, "nvidia_rtx_5080": 56.0, "nvidia_rtx_5090": 104.5}, "model_id": "mistral/voxtral-small-24b-2507", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Mistral AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "moondream1", "fits": {"amd_rx_7900_xt": 700.0, "amd_rx_7900_xtx": 840.0, "amd_rx_9070_xt": 564.4, "apple_m1_max": 350.0, "apple_m2_max": 350.0, "apple_m2_ultra": 700.0, "apple_m3_max": 350.0, "apple_m3_ultra": 716.6, "apple_m4": 105.0, "apple_m4_max": 477.7, "apple_m4_pro": 238.9, "nvidia_rtx_3060_12gb": 315.0, "nvidia_rtx_3090": 819.0, "nvidia_rtx_4060_ti_16gb": 252.0, "nvidia_rtx_4070_ti_super": 588.0, "nvidia_rtx_4080_super": 644.0, "nvidia_rtx_4090": 882.0, "nvidia_rtx_5080": 840.0, "nvidia_rtx_5090": 1568.0}, "model_id": "moondream/moondream1", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Moondream", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "moondream2", "fits": {"amd_rx_7900_xt": 589.5, "amd_rx_7900_xtx": 707.4, "amd_rx_9070_xt": 475.3, "apple_m1_max": 294.7, "apple_m2_max": 294.7, "apple_m2_ultra": 589.5, "apple_m3_max": 294.7, "apple_m3_ultra": 603.5, "apple_m4": 88.4, "apple_m4_max": 402.3, "apple_m4_pro": 201.2, "nvidia_rtx_3060_12gb": 265.3, "nvidia_rtx_3090": 689.7, "nvidia_rtx_4060_ti_16gb": 212.2, "nvidia_rtx_4070_ti_super": 495.2, "nvidia_rtx_4080_super": 542.3, "nvidia_rtx_4090": 742.7, "nvidia_rtx_5080": 707.4, "nvidia_rtx_5090": 1320.4}, "model_id": "moondream/moondream2", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Moondream", "scores_as_of": null}, {"benchmark_scores": {"gsm8k": 97.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.6, "display_name": "Kimi K2 0711", "fits": {}, "model_id": "moonshot/kimi-k2-0711-preview", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Moonshot AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 97.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.6, "display_name": "Kimi K2 0905", "fits": {}, "model_id": "moonshot/kimi-k2-0905-preview", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Moonshot AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 97.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.6, "display_name": "Kimi K2.5", "fits": {}, "model_id": "moonshot/kimi-k2-5", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Moonshot AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 97.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 1.15, "display_name": "Kimi K2 Thinking Turbo", "fits": {}, "model_id": "moonshot/kimi-k2-thinking-turbo", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Moonshot AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 97.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.6, "display_name": "Kimi K2 Thinking", "fits": {}, "model_id": "moonshot/kimi-k2-thinking", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Moonshot AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"gsm8k": 97.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 2.4, "display_name": "Kimi K2 Turbo", "fits": {}, "model_id": "moonshot/kimi-k2-turbo-preview", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Moonshot AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"mteb_pair_classification": 81.0, "mteb_reranking": 53.0, "mteb_sts": 78.0, "mteb_summarization": 27.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "CodeRankEmbed", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/coderankembed", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "CodeRankLLM", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nomic/coderankllm", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Nomic AI", "scores_as_of": null}, {"benchmark_scores": {"miracl": 65.8, "mteb_pair_classification": 84.0, "mteb_reranking": 56.5, "mteb_sts": 82.5, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "colnomic embed multimodal 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "nomic/colnomic-embed-multimodal-7b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "gpt4all j", "fits": {"amd_rx_7900_xt": 167.2, "amd_rx_7900_xtx": 200.6, "amd_rx_9070_xt": 134.8, "apple_m1_max": 83.6, "apple_m2_max": 83.6, "apple_m2_ultra": 167.2, "apple_m3_max": 83.6, "apple_m3_ultra": 171.1, "apple_m4": 25.1, "apple_m4_max": 114.1, "apple_m4_pro": 57.0, "nvidia_rtx_3060_12gb": 75.2, "nvidia_rtx_3090": 195.6, "nvidia_rtx_4060_ti_16gb": 60.2, "nvidia_rtx_4070_ti_super": 140.4, "nvidia_rtx_4080_super": 153.8, "nvidia_rtx_4090": 210.6, "nvidia_rtx_5080": 200.6, "nvidia_rtx_5090": 374.4}, "model_id": "nomic/gpt4all-j", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nomic AI", "scores_as_of": null}, {"benchmark_scores": {"miracl": 50.0, "mteb_pair_classification": 83.0, "mteb_reranking": 55.2, "mteb_sts": 81.0, "mteb_summarization": 28.8}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "modernbert embed base", "fits": {"amd_rx_7900_xt": 7516.8, "amd_rx_7900_xtx": 9020.1, "amd_rx_9070_xt": 6060.4, "apple_m1_max": 3758.4, "apple_m2_max": 3758.4, "apple_m2_ultra": 7516.8, "apple_m3_max": 3758.4, "apple_m3_ultra": 7695.3, "apple_m4": 1127.5, "apple_m4_max": 5130.2, "apple_m4_pro": 2565.1, "nvidia_rtx_3060_12gb": 3382.6, "nvidia_rtx_3090": 8794.6, "nvidia_rtx_4060_ti_16gb": 2706.0, "nvidia_rtx_4070_ti_super": 6314.1, "nvidia_rtx_4080_super": 6915.4, "nvidia_rtx_4090": 9471.1, "nvidia_rtx_5080": 9020.1, "nvidia_rtx_5090": 16837.6}, "model_id": "nomic/modernbert-embed-base", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "nomic bert 2048", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/nomic-bert-2048", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nomic AI", "scores_as_of": null}, {"benchmark_scores": {"mteb_classification": 66.2, "mteb_clustering": 44.0, "mteb_overall": 60.5, "mteb_pair_classification": 81.0, "mteb_reranking": 53.0, "mteb_retrieval": 57.0, "mteb_sts": 78.0, "mteb_summarization": 27.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed code", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/nomic-embed-code", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 65.8, "mteb_pair_classification": 84.0, "mteb_reranking": 56.5, "mteb_sts": 82.5, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed multimodal 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "nomic/nomic-embed-multimodal-7b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 52.3, "miracl": 50.0, "mteb_classification": 68.8, "mteb_clustering": 46.8, "mteb_overall": 62.8, "mteb_pair_classification": 83.0, "mteb_reranking": 55.2, "mteb_retrieval": 59.1, "mteb_sts": 81.0, "mteb_summarization": 28.8}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed text v1.5 GGUF", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/nomic-embed-text-v1-5-gguf", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 52.3, "miracl": 50.0, "mteb_classification": 68.8, "mteb_clustering": 46.8, "mteb_overall": 62.8, "mteb_pair_classification": 83.0, "mteb_reranking": 55.2, "mteb_retrieval": 59.1, "mteb_sts": 81.0, "mteb_summarization": 28.8}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed text v1.5", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/nomic-embed-text-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 50.8, "miracl": 47.5, "mteb_classification": 67.5, "mteb_clustering": 45.5, "mteb_overall": 61.5, "mteb_pair_classification": 82.5, "mteb_reranking": 54.5, "mteb_retrieval": 57.8, "mteb_sts": 80.2, "mteb_summarization": 28.2}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed text v1", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/nomic-embed-text-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 55.0, "miracl": 65.8, "mteb_classification": 70.5, "mteb_clustering": 49.5, "mteb_overall": 65.5, "mteb_pair_classification": 84.0, "mteb_reranking": 56.5, "mteb_retrieval": 62.2, "mteb_sts": 82.5, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed text v2 moe GGUF", "fits": {"amd_rx_7900_xt": 2357.9, "amd_rx_7900_xtx": 2829.5, "amd_rx_9070_xt": 1901.1, "apple_m1_max": 1178.9, "apple_m2_max": 1178.9, "apple_m2_ultra": 2357.9, "apple_m3_max": 1178.9, "apple_m3_ultra": 2413.9, "apple_m4": 353.7, "apple_m4_max": 1609.3, "apple_m4_pro": 804.6, "nvidia_rtx_3060_12gb": 1061.1, "nvidia_rtx_3090": 2758.7, "nvidia_rtx_4060_ti_16gb": 848.8, "nvidia_rtx_4070_ti_super": 1980.6, "nvidia_rtx_4080_super": 2169.3, "nvidia_rtx_4090": 2970.9, "nvidia_rtx_5080": 2829.5, "nvidia_rtx_5090": 5281.7}, "model_id": "nomic/nomic-embed-text-v2-moe-gguf", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 55.0, "miracl": 65.8, "mteb_classification": 70.5, "mteb_clustering": 49.5, "mteb_overall": 65.5, "mteb_pair_classification": 84.0, "mteb_reranking": 56.5, "mteb_retrieval": 62.2, "mteb_sts": 82.5, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed text v2 moe", "fits": {"amd_rx_7900_xt": 2357.9, "amd_rx_7900_xtx": 2829.5, "amd_rx_9070_xt": 1901.1, "apple_m1_max": 1178.9, "apple_m2_max": 1178.9, "apple_m2_ultra": 2357.9, "apple_m3_max": 1178.9, "apple_m3_ultra": 2413.9, "apple_m4": 353.7, "apple_m4_max": 1609.3, "apple_m4_pro": 804.6, "nvidia_rtx_3060_12gb": 1061.1, "nvidia_rtx_3090": 2758.7, "nvidia_rtx_4060_ti_16gb": 848.8, "nvidia_rtx_4070_ti_super": 1980.6, "nvidia_rtx_4080_super": 2169.3, "nvidia_rtx_4090": 2970.9, "nvidia_rtx_5080": 2829.5, "nvidia_rtx_5090": 5281.7}, "model_id": "nomic/nomic-embed-text-v2-moe", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed vision v1.5", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/nomic-embed-vision-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "nomic embed vision v1", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "nomic/nomic-embed-vision-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Nomic AI", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 63.5, "bbh": 50.7, "gpqa_diamond": 29.3, "gsm8k": 67.9, "hellaswag": 83.2, "ifeval": 53.6, "math_500": 8.4, "mmlu_abstract_algebra": 36.0, "mmlu_anatomy": 63.0, "mmlu_astronomy": 69.7, "mmlu_business_ethics": 63.0, "mmlu_clinical_knowledge": 73.6, "mmlu_college_biology": 72.2, "mmlu_college_chemistry": 41.0, "mmlu_college_computer_science": 47.0, "mmlu_college_mathematics": 36.0, "mmlu_college_medicine": 61.8, "mmlu_college_physics": 46.1, "mmlu_computer_security": 74.0, "mmlu_conceptual_physics": 56.6, "mmlu_econometrics": 50.9, "mmlu_electrical_engineering": 60.0, "mmlu_elementary_mathematics": 42.9, "mmlu_formal_logic": 47.6, "mmlu_global_facts": 42.0, "mmlu_high_school_biology": 80.6, "mmlu_high_school_chemistry": 47.8, "mmlu_high_school_computer_science": 66.0, "mmlu_high_school_european_history": 74.5, "mmlu_high_school_geography": 78.8, "mmlu_high_school_government_and_politics": 88.1, "mmlu_high_school_macroeconomics": 62.1, "mmlu_high_school_mathematics": 38.5, "mmlu_high_school_microeconomics": 72.7, "mmlu_high_school_physics": 39.1, "mmlu_high_school_psychology": 82.9, "mmlu_high_school_statistics": 50.9, "mmlu_high_school_us_history": 81.4, "mmlu_high_school_world_history": 84.8, "mmlu_human_aging": 70.4, "mmlu_human_sexuality": 76.3, "mmlu_international_law": 78.5, "mmlu_jurisprudence": 73.1, "mmlu_logical_fallacies": 71.8, "mmlu_machine_learning": 54.5, "mmlu_management": 80.6, "mmlu_marketing": 87.6, "mmlu_medical_genetics": 78.0, "mmlu_miscellaneous": 83.1, "mmlu_moral_disputes": 71.1, "mmlu_moral_scenarios": 43.7, "mmlu_nutrition": 71.2, "mmlu_philosophy": 71.1, "mmlu_prehistory": 71.3, "mmlu_pro": 30.5, "mmlu_professional_accounting": 50.7, "mmlu_professional_law": 46.8, "mmlu_professional_medicine": 69.5, "mmlu_professional_psychology": 69.8, "mmlu_public_relations": 61.8, "mmlu_security_studies": 75.5, "mmlu_sociology": 86.6, "mmlu_us_foreign_policy": 84.0, "mmlu_virology": 53.0, "mmlu_world_religions": 81.9, "musr": 42.6, "truthfulqa": 56.6, "winogrande": 76.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Hermes 2 Pro Llama 3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/hermes-2-pro-llama-3-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 63.1, "bbh": 52.1, "gpqa_diamond": 30.4, "gsm8k": 70.4, "hellaswag": 82.6, "ifeval": 65.2, "math_500": 9.7, "mmlu_abstract_algebra": 38.0, "mmlu_anatomy": 63.7, "mmlu_astronomy": 69.1, "mmlu_business_ethics": 64.0, "mmlu_clinical_knowledge": 74.3, "mmlu_college_biology": 74.3, "mmlu_college_chemistry": 48.0, "mmlu_college_computer_science": 49.0, "mmlu_college_mathematics": 35.0, "mmlu_college_medicine": 66.5, "mmlu_college_physics": 48.0, "mmlu_computer_security": 79.0, "mmlu_conceptual_physics": 55.7, "mmlu_econometrics": 55.3, "mmlu_electrical_engineering": 61.4, "mmlu_elementary_mathematics": 45.2, "mmlu_formal_logic": 48.4, "mmlu_global_facts": 39.0, "mmlu_high_school_biology": 81.3, "mmlu_high_school_chemistry": 51.7, "mmlu_high_school_computer_science": 70.0, "mmlu_high_school_european_history": 75.8, "mmlu_high_school_geography": 81.8, "mmlu_high_school_government_and_politics": 88.6, "mmlu_high_school_macroeconomics": 65.1, "mmlu_high_school_mathematics": 38.1, "mmlu_high_school_microeconomics": 75.2, "mmlu_high_school_physics": 45.0, "mmlu_high_school_psychology": 83.1, "mmlu_high_school_statistics": 50.9, "mmlu_high_school_us_history": 83.8, "mmlu_high_school_world_history": 86.9, "mmlu_human_aging": 72.6, "mmlu_human_sexuality": 80.9, "mmlu_international_law": 78.5, "mmlu_jurisprudence": 75.9, "mmlu_logical_fallacies": 73.6, "mmlu_machine_learning": 55.4, "mmlu_management": 80.6, "mmlu_marketing": 89.3, "mmlu_medical_genetics": 79.0, "mmlu_miscellaneous": 84.4, "mmlu_moral_disputes": 74.9, "mmlu_moral_scenarios": 43.9, "mmlu_nutrition": 73.5, "mmlu_philosophy": 73.6, "mmlu_prehistory": 73.8, "mmlu_pro": 33.7, "mmlu_professional_accounting": 52.1, "mmlu_professional_law": 48.5, "mmlu_professional_medicine": 69.5, "mmlu_professional_psychology": 70.3, "mmlu_public_relations": 65.5, "mmlu_security_studies": 77.1, "mmlu_sociology": 87.6, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 54.8, "mmlu_world_religions": 80.7, "musr": 39.5, "truthfulqa": 55.7, "winogrande": 76.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Hermes 2 Theta Llama 3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/hermes-2-theta-llama-3-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0, "bbh": 51.8, "gpqa_diamond": 29.8, "ifeval": 61.7, "math_500": 4.8, "mmlu_pro": 31.4, "musr": 43.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Hermes 3 Llama 3.1 8B GGUF", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/hermes-3-llama-3-1-8b-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2025-03"}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0, "bbh": 51.8, "gpqa_diamond": 29.8, "ifeval": 61.7, "math_500": 4.8, "mmlu_pro": 31.4, "musr": 43.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Hermes 3 Llama 3.1 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/hermes-3-llama-3-1-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 43.5, "gpqa_diamond": 27.5, "ifeval": 38.2, "math_500": 3.9, "mmlu_pro": 25.4, "musr": 40.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.06, "display_name": "Hermes 3 Llama 3.2 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "nous-research/hermes-3-llama-3-2-3b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hermes 4.3 36B", "fits": {"amd_rx_7900_xtx": 37.3, "apple_m1_max": 15.6, "apple_m2_max": 15.6, "apple_m2_ultra": 31.1, "apple_m3_max": 15.6, "apple_m3_ultra": 31.8, "apple_m4": 4.7, "apple_m4_max": 21.2, "apple_m4_pro": 10.6, "nvidia_rtx_3090": 36.4, "nvidia_rtx_4090": 39.2, "nvidia_rtx_5090": 69.7}, "model_id": "nous-research/hermes-4-3-36b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hermes 4 70B FP8", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "nous-research/hermes-4-70b-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": null}, {"benchmark_scores": {"bbh": 41.3, "gpqa_diamond": 28.1, "gsm8k": 22.8, "ifeval": 24.8, "math_500": 1.5, "mmlu_pro": 23.8, "musr": 35.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 13B hf", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "nous-research/llama-2-13b-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 31.1, "gpqa_diamond": 25.3, "gsm8k": 7.4, "ifeval": 39.9, "math_500": 2.0, "mmlu_pro": 16.9, "musr": 36.8, "winogrande": 71.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 7B chat hf", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "nous-research/llama-2-7b-chat-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 35.0, "gpqa_diamond": 26.7, "gsm8k": 14.5, "ifeval": 25.2, "math_500": 1.7, "mmlu_pro": 18.6, "musr": 37.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Llama 2 7B hf", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "nous-research/llama-2-7b-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 31.1, "gpqa_diamond": 22.8, "ifeval": 14.8, "math_500": 1.2, "mmlu_pro": 12.0, "musr": 34.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.04, "display_name": "Llama 3.2 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "nous-research/llama-3-2-1b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2025-03"}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Meta Llama 3.1 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/meta-llama-3-1-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": null}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Meta Llama 3.1 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/meta-llama-3-1-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 71.4, "bbh": 65.5, "gpqa_diamond": 28.7, "gsm8k": 95.0, "hellaswag": 85.7, "ifeval": 81.0, "math_500": 24.5, "mgsm": 91.1, "mmlu_abstract_algebra": 51.0, "mmlu_anatomy": 77.0, "mmlu_astronomy": 93.4, "mmlu_business_ethics": 86.0, "mmlu_clinical_knowledge": 86.8, "mmlu_college_biology": 91.0, "mmlu_college_chemistry": 56.0, "mmlu_college_computer_science": 69.0, "mmlu_college_mathematics": 58.0, "mmlu_college_medicine": 81.5, "mmlu_college_physics": 59.8, "mmlu_computer_security": 82.0, "mmlu_conceptual_physics": 81.7, "mmlu_econometrics": 72.8, "mmlu_electrical_engineering": 77.9, "mmlu_elementary_mathematics": 70.9, "mmlu_formal_logic": 62.7, "mmlu_global_facts": 59.0, "mmlu_high_school_biology": 89.7, "mmlu_high_school_chemistry": 69.0, "mmlu_high_school_computer_science": 92.0, "mmlu_high_school_european_history": 86.7, "mmlu_high_school_geography": 92.9, "mmlu_high_school_government_and_politics": 98.4, "mmlu_high_school_macroeconomics": 83.8, "mmlu_high_school_mathematics": 54.4, "mmlu_high_school_microeconomics": 88.7, "mmlu_high_school_physics": 62.3, "mmlu_high_school_psychology": 94.1, "mmlu_high_school_statistics": 73.6, "mmlu_high_school_us_history": 92.2, "mmlu_high_school_world_history": 92.8, "mmlu_human_aging": 82.1, "mmlu_human_sexuality": 87.0, "mmlu_international_law": 92.6, "mmlu_jurisprudence": 88.0, "mmlu_logical_fallacies": 85.9, "mmlu_machine_learning": 70.5, "mmlu_management": 89.3, "mmlu_marketing": 93.6, "mmlu_medical_genetics": 92.0, "mmlu_miscellaneous": 91.8, "mmlu_moral_disputes": 84.4, "mmlu_moral_scenarios": 70.8, "mmlu_nutrition": 87.6, "mmlu_philosophy": 81.4, "mmlu_prehistory": 90.1, "mmlu_pro": 52.1, "mmlu_professional_accounting": 63.8, "mmlu_professional_law": 64.1, "mmlu_professional_medicine": 89.0, "mmlu_professional_psychology": 85.3, "mmlu_public_relations": 74.5, "mmlu_security_studies": 81.6, "mmlu_sociology": 92.0, "mmlu_us_foreign_policy": 92.0, "mmlu_virology": 56.6, "mmlu_world_religions": 90.1, "musr": 41.5, "truthfulqa": 61.8, "winogrande": 82.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Meta Llama 3 70B Instruct", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "nous-research/meta-llama-3-70b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arc_challenge": 60.8, "bbh": 49.1, "gpqa_diamond": 29.3, "gsm8k": 68.7, "hellaswag": 78.6, "ifeval": 47.8, "math_500": 9.1, "mmlu_abstract_algebra": 32.0, "mmlu_anatomy": 65.2, "mmlu_astronomy": 70.4, "mmlu_business_ethics": 69.0, "mmlu_clinical_knowledge": 74.7, "mmlu_college_biology": 79.9, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 59.0, "mmlu_college_mathematics": 39.0, "mmlu_college_medicine": 63.6, "mmlu_college_physics": 50.0, "mmlu_computer_security": 77.0, "mmlu_conceptual_physics": 60.4, "mmlu_econometrics": 60.5, "mmlu_electrical_engineering": 62.8, "mmlu_elementary_mathematics": 44.4, "mmlu_formal_logic": 48.4, "mmlu_global_facts": 40.0, "mmlu_high_school_biology": 78.4, "mmlu_high_school_chemistry": 50.7, "mmlu_high_school_computer_science": 75.0, "mmlu_high_school_european_history": 74.5, "mmlu_high_school_geography": 84.3, "mmlu_high_school_government_and_politics": 91.2, "mmlu_high_school_macroeconomics": 65.9, "mmlu_high_school_mathematics": 40.0, "mmlu_high_school_microeconomics": 76.5, "mmlu_high_school_physics": 44.4, "mmlu_high_school_psychology": 83.7, "mmlu_high_school_statistics": 53.7, "mmlu_high_school_us_history": 85.8, "mmlu_high_school_world_history": 84.4, "mmlu_human_aging": 72.2, "mmlu_human_sexuality": 77.9, "mmlu_international_law": 81.8, "mmlu_jurisprudence": 77.8, "mmlu_logical_fallacies": 76.7, "mmlu_machine_learning": 54.5, "mmlu_management": 78.6, "mmlu_marketing": 90.6, "mmlu_medical_genetics": 80.0, "mmlu_miscellaneous": 79.8, "mmlu_moral_disputes": 74.9, "mmlu_moral_scenarios": 43.7, "mmlu_nutrition": 74.5, "mmlu_philosophy": 71.4, "mmlu_prehistory": 74.1, "mmlu_pro": 35.9, "mmlu_professional_accounting": 52.8, "mmlu_professional_law": 47.8, "mmlu_professional_medicine": 71.7, "mmlu_professional_psychology": 70.6, "mmlu_public_relations": 64.5, "mmlu_security_studies": 73.9, "mmlu_sociology": 86.6, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 51.2, "mmlu_world_religions": 77.8, "musr": 38.1, "truthfulqa": 51.6, "winogrande": 74.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Meta Llama 3 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/meta-llama-3-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 60.2, "bbh": 46.0, "gpqa_diamond": 30.5, "gsm8k": 45.2, "hellaswag": 82.2, "ifeval": 14.6, "math_500": 4.5, "mmlu_abstract_algebra": 35.0, "mmlu_anatomy": 69.6, "mmlu_astronomy": 67.8, "mmlu_business_ethics": 63.0, "mmlu_clinical_knowledge": 75.5, "mmlu_college_biology": 77.1, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 49.0, "mmlu_college_mathematics": 33.0, "mmlu_college_medicine": 63.6, "mmlu_college_physics": 49.0, "mmlu_computer_security": 80.0, "mmlu_conceptual_physics": 59.1, "mmlu_econometrics": 49.1, "mmlu_electrical_engineering": 64.8, "mmlu_elementary_mathematics": 43.4, "mmlu_formal_logic": 47.6, "mmlu_global_facts": 37.0, "mmlu_high_school_biology": 76.8, "mmlu_high_school_chemistry": 54.7, "mmlu_high_school_computer_science": 68.0, "mmlu_high_school_european_history": 77.0, "mmlu_high_school_geography": 80.3, "mmlu_high_school_government_and_politics": 89.1, "mmlu_high_school_macroeconomics": 65.1, "mmlu_high_school_mathematics": 40.0, "mmlu_high_school_microeconomics": 73.5, "mmlu_high_school_physics": 44.4, "mmlu_high_school_psychology": 84.8, "mmlu_high_school_statistics": 57.9, "mmlu_high_school_us_history": 85.8, "mmlu_high_school_world_history": 82.3, "mmlu_human_aging": 71.7, "mmlu_human_sexuality": 77.1, "mmlu_international_law": 85.1, "mmlu_jurisprudence": 75.0, "mmlu_logical_fallacies": 73.6, "mmlu_machine_learning": 53.6, "mmlu_management": 87.4, "mmlu_marketing": 89.7, "mmlu_medical_genetics": 79.0, "mmlu_miscellaneous": 83.0, "mmlu_moral_disputes": 74.0, "mmlu_moral_scenarios": 41.3, "mmlu_nutrition": 76.5, "mmlu_philosophy": 74.3, "mmlu_prehistory": 72.5, "mmlu_pro": 32.1, "mmlu_professional_accounting": 49.3, "mmlu_professional_law": 46.7, "mmlu_professional_medicine": 72.1, "mmlu_professional_psychology": 72.4, "mmlu_public_relations": 71.8, "mmlu_security_studies": 75.1, "mmlu_sociology": 85.6, "mmlu_us_foreign_policy": 86.0, "mmlu_virology": 56.0, "mmlu_world_religions": 83.0, "musr": 36.1, "truthfulqa": 42.9, "winogrande": 78.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Meta Llama 3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nous-research/meta-llama-3-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 71.1, "bbh": 55.4, "gpqa_diamond": 32.1, "gsm8k": 71.6, "hellaswag": 87.3, "ifeval": 59.0, "math_500": 12.2, "mmlu_abstract_algebra": 43.0, "mmlu_anatomy": 68.9, "mmlu_astronomy": 80.3, "mmlu_business_ethics": 73.0, "mmlu_clinical_knowledge": 79.2, "mmlu_college_biology": 83.3, "mmlu_college_chemistry": 55.0, "mmlu_college_computer_science": 66.0, "mmlu_college_mathematics": 43.0, "mmlu_college_medicine": 69.4, "mmlu_college_physics": 49.0, "mmlu_computer_security": 81.0, "mmlu_conceptual_physics": 69.4, "mmlu_econometrics": 64.9, "mmlu_electrical_engineering": 69.0, "mmlu_elementary_mathematics": 50.3, "mmlu_formal_logic": 57.1, "mmlu_global_facts": 45.0, "mmlu_high_school_biology": 85.2, "mmlu_high_school_chemistry": 59.1, "mmlu_high_school_computer_science": 75.0, "mmlu_high_school_european_history": 78.2, "mmlu_high_school_geography": 85.4, "mmlu_high_school_government_and_politics": 93.8, "mmlu_high_school_macroeconomics": 69.0, "mmlu_high_school_mathematics": 34.8, "mmlu_high_school_microeconomics": 79.4, "mmlu_high_school_physics": 45.7, "mmlu_high_school_psychology": 89.5, "mmlu_high_school_statistics": 66.2, "mmlu_high_school_us_history": 86.8, "mmlu_high_school_world_history": 89.5, "mmlu_human_aging": 76.2, "mmlu_human_sexuality": 87.8, "mmlu_international_law": 86.0, "mmlu_jurisprudence": 84.3, "mmlu_logical_fallacies": 80.4, "mmlu_machine_learning": 54.5, "mmlu_management": 85.4, "mmlu_marketing": 90.2, "mmlu_medical_genetics": 80.0, "mmlu_miscellaneous": 88.0, "mmlu_moral_disputes": 80.1, "mmlu_moral_scenarios": 57.2, "mmlu_nutrition": 80.7, "mmlu_philosophy": 79.1, "mmlu_prehistory": 85.5, "mmlu_pro": 36.7, "mmlu_professional_accounting": 52.8, "mmlu_professional_law": 55.5, "mmlu_professional_medicine": 78.7, "mmlu_professional_psychology": 78.8, "mmlu_public_relations": 69.1, "mmlu_security_studies": 81.2, "mmlu_sociology": 87.1, "mmlu_us_foreign_policy": 91.0, "mmlu_virology": 50.6, "mmlu_world_religions": 89.5, "musr": 46.0, "truthfulqa": 54.8, "winogrande": 83.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.24, "display_name": "Nous Hermes 2 Mixtral 8x7B DPO", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "nous-research/nous-hermes-2-mixtral-8x7b-dpo", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 66.7, "bbh": 54.1, "gpqa_diamond": 29.4, "gsm8k": 69.4, "hellaswag": 84.9, "ifeval": 52.8, "math_500": 5.7, "mmlu_abstract_algebra": 35.0, "mmlu_anatomy": 56.3, "mmlu_astronomy": 76.3, "mmlu_business_ethics": 74.0, "mmlu_clinical_knowledge": 69.1, "mmlu_college_biology": 72.9, "mmlu_college_chemistry": 47.0, "mmlu_college_computer_science": 42.0, "mmlu_college_mathematics": 37.0, "mmlu_college_medicine": 64.2, "mmlu_college_physics": 41.2, "mmlu_computer_security": 73.0, "mmlu_conceptual_physics": 59.6, "mmlu_econometrics": 54.4, "mmlu_electrical_engineering": 57.9, "mmlu_elementary_mathematics": 48.9, "mmlu_formal_logic": 45.2, "mmlu_global_facts": 32.0, "mmlu_high_school_biology": 80.6, "mmlu_high_school_chemistry": 51.7, "mmlu_high_school_computer_science": 71.0, "mmlu_high_school_european_history": 83.0, "mmlu_high_school_geography": 87.9, "mmlu_high_school_government_and_politics": 89.6, "mmlu_high_school_macroeconomics": 67.4, "mmlu_high_school_mathematics": 36.3, "mmlu_high_school_microeconomics": 69.3, "mmlu_high_school_physics": 37.7, "mmlu_high_school_psychology": 85.1, "mmlu_high_school_statistics": 52.8, "mmlu_high_school_us_history": 83.8, "mmlu_high_school_world_history": 87.8, "mmlu_human_aging": 74.9, "mmlu_human_sexuality": 77.9, "mmlu_international_law": 81.8, "mmlu_jurisprudence": 78.7, "mmlu_logical_fallacies": 74.8, "mmlu_machine_learning": 50.9, "mmlu_management": 80.6, "mmlu_marketing": 88.0, "mmlu_medical_genetics": 74.0, "mmlu_miscellaneous": 82.8, "mmlu_moral_disputes": 74.0, "mmlu_moral_scenarios": 34.9, "mmlu_nutrition": 77.8, "mmlu_philosophy": 72.7, "mmlu_prehistory": 77.5, "mmlu_pro": 34.6, "mmlu_professional_accounting": 51.8, "mmlu_professional_law": 50.1, "mmlu_professional_medicine": 75.7, "mmlu_professional_psychology": 68.5, "mmlu_public_relations": 70.0, "mmlu_security_studies": 78.8, "mmlu_sociology": 82.6, "mmlu_us_foreign_policy": 91.0, "mmlu_virology": 56.0, "mmlu_world_religions": 83.0, "musr": 43.7, "truthfulqa": 55.8, "winogrande": 82.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Nous Hermes 2 SOLAR 10.7B", "fits": {"amd_rx_7900_xt": 104.7, "amd_rx_7900_xtx": 125.6, "amd_rx_9070_xt": 84.4, "apple_m1_max": 52.3, "apple_m2_max": 52.3, "apple_m2_ultra": 104.7, "apple_m3_max": 52.3, "apple_m3_ultra": 107.2, "apple_m4": 15.7, "apple_m4_max": 71.4, "apple_m4_pro": 35.7, "nvidia_rtx_3060_12gb": 47.1, "nvidia_rtx_3090": 122.5, "nvidia_rtx_4060_ti_16gb": 37.7, "nvidia_rtx_4070_ti_super": 87.9, "nvidia_rtx_4080_super": 96.3, "nvidia_rtx_4090": 131.9, "nvidia_rtx_5080": 125.6, "nvidia_rtx_5090": 234.5}, "model_id": "nous-research/nous-hermes-2-solar-10-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 66.9, "gsm8k": 70.1, "hellaswag": 85.5, "mmlu_abstract_algebra": 49.0, "mmlu_anatomy": 71.9, "mmlu_astronomy": 89.5, "mmlu_business_ethics": 78.0, "mmlu_clinical_knowledge": 80.0, "mmlu_college_biology": 90.3, "mmlu_college_chemistry": 51.0, "mmlu_college_computer_science": 65.0, "mmlu_college_mathematics": 49.0, "mmlu_college_medicine": 69.9, "mmlu_college_physics": 52.9, "mmlu_computer_security": 83.0, "mmlu_conceptual_physics": 79.1, "mmlu_econometrics": 57.0, "mmlu_electrical_engineering": 77.2, "mmlu_elementary_mathematics": 69.0, "mmlu_formal_logic": 57.9, "mmlu_global_facts": 51.0, "mmlu_high_school_biology": 89.7, "mmlu_high_school_chemistry": 62.1, "mmlu_high_school_computer_science": 84.0, "mmlu_high_school_european_history": 87.9, "mmlu_high_school_geography": 89.9, "mmlu_high_school_government_and_politics": 97.9, "mmlu_high_school_macroeconomics": 82.8, "mmlu_high_school_mathematics": 41.5, "mmlu_high_school_microeconomics": 85.3, "mmlu_high_school_physics": 50.3, "mmlu_high_school_psychology": 92.1, "mmlu_high_school_statistics": 66.2, "mmlu_high_school_us_history": 91.2, "mmlu_high_school_world_history": 90.7, "mmlu_human_aging": 79.4, "mmlu_human_sexuality": 89.3, "mmlu_international_law": 90.9, "mmlu_jurisprudence": 89.8, "mmlu_logical_fallacies": 87.1, "mmlu_machine_learning": 60.7, "mmlu_management": 92.2, "mmlu_marketing": 91.9, "mmlu_medical_genetics": 86.0, "mmlu_miscellaneous": 91.1, "mmlu_moral_disputes": 83.5, "mmlu_moral_scenarios": 71.1, "mmlu_nutrition": 84.3, "mmlu_philosophy": 81.4, "mmlu_prehistory": 88.9, "mmlu_professional_accounting": 64.9, "mmlu_professional_law": 61.7, "mmlu_professional_medicine": 83.1, "mmlu_professional_psychology": 82.4, "mmlu_public_relations": 71.8, "mmlu_security_studies": 84.1, "mmlu_sociology": 87.6, "mmlu_us_foreign_policy": 92.0, "mmlu_virology": 57.2, "mmlu_world_religions": 87.1, "truthfulqa": 60.4, "winogrande": 83.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Nous Hermes 2 Yi 34B", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "nous-research/nous-hermes-2-yi-34b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 38.2, "gpqa_diamond": 26.3, "gsm8k": 5.8, "ifeval": 17.3, "math_500": 0.9, "mmlu_pro": 19.4, "musr": 42.6, "winogrande": 74.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Nous Hermes llama 2 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "nous-research/nous-hermes-llama-2-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Nous Research", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "audio flamingo 3 hf", "fits": {}, "model_id": "nvidia/audio-flamingo-3-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "bigvgan v2 22khz 80band 256x", "fits": {"amd_rx_7900_xt": 10000.0, "amd_rx_7900_xtx": 12000.0, "amd_rx_9070_xt": 8062.5, "apple_m1_max": 5000.0, "apple_m2_max": 5000.0, "apple_m2_ultra": 10000.0, "apple_m3_max": 5000.0, "apple_m3_ultra": 10237.5, "apple_m4": 1500.0, "apple_m4_max": 6825.0, "apple_m4_pro": 3412.5, "nvidia_rtx_3060_12gb": 4500.0, "nvidia_rtx_3090": 11700.0, "nvidia_rtx_4060_ti_16gb": 3600.0, "nvidia_rtx_4070_ti_super": 8400.0, "nvidia_rtx_4080_super": 9200.0, "nvidia_rtx_4090": 12600.0, "nvidia_rtx_5080": 12000.0, "nvidia_rtx_5090": 22400.0}, "model_id": "nvidia/bigvgan-v2-22khz-80band-256x", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "bigvgan v2 44khz 128band 512x", "fits": {"amd_rx_7900_xt": 9180.3, "amd_rx_7900_xtx": 11016.4, "amd_rx_9070_xt": 7401.6, "apple_m1_max": 4590.2, "apple_m2_max": 4590.2, "apple_m2_ultra": 9180.3, "apple_m3_max": 4590.2, "apple_m3_ultra": 9398.4, "apple_m4": 1377.0, "apple_m4_max": 6265.6, "apple_m4_pro": 3132.8, "nvidia_rtx_3060_12gb": 4131.1, "nvidia_rtx_3090": 10741.0, "nvidia_rtx_4060_ti_16gb": 3304.9, "nvidia_rtx_4070_ti_super": 7711.5, "nvidia_rtx_4080_super": 8445.9, "nvidia_rtx_4090": 11567.2, "nvidia_rtx_5080": 11016.4, "nvidia_rtx_5090": 20563.9}, "model_id": "nvidia/bigvgan-v2-44khz-128band-512x", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "canary 1B flash", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "nvidia/canary-1b-flash", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "canary 1B v2", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "nvidia/canary-1b-v2", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "canary qwen 2.5B", "fits": {"amd_rx_7900_xt": 448.0, "amd_rx_7900_xtx": 537.6, "amd_rx_9070_xt": 361.2, "apple_m1_max": 224.0, "apple_m2_max": 224.0, "apple_m2_ultra": 448.0, "apple_m3_max": 224.0, "apple_m3_ultra": 458.6, "apple_m4": 67.2, "apple_m4_max": 305.8, "apple_m4_pro": 152.9, "nvidia_rtx_3060_12gb": 201.6, "nvidia_rtx_3090": 524.2, "nvidia_rtx_4060_ti_16gb": 161.3, "nvidia_rtx_4070_ti_super": 376.3, "nvidia_rtx_4080_super": 412.2, "nvidia_rtx_4090": 564.5, "nvidia_rtx_5080": 537.6, "nvidia_rtx_5090": 1003.5}, "model_id": "nvidia/canary-qwen-2-5b", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {"aider_polyglot": 65.8, "aime_2025": 82.5, "alpaca_eval": 45.2, "arena_elo_coding": 1380.0, "arena_elo_math": 1410.0, "arena_elo_overall": 1370.0, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 40.0, "bbh": 68.7, "gpqa_diamond": 73.2, "humaneval": 90.5, "ifeval": 88.0, "live_code_bench": 77.0, "math_500": 97.5, "mmlu_astronomy": 76.5, "mmlu_biology": 84.2, "mmlu_business_ethics": 77.8, "mmlu_chemistry": 78.5, "mmlu_clinical_knowledge": 83.5, "mmlu_computer_science": 85.5, "mmlu_jurisprudence": 74.5, "mmlu_physics": 82.8, "mmlu_pro": 81.0, "mmlu_professional_accounting": 68.2, "mmlu_professional_law": 72.5, "mt_bench": 9.0, "multipl_e": 84.6, "multipl_e_cpp": 83.5, "multipl_e_csharp": 82.1, "multipl_e_go": 80.5, "multipl_e_java": 87.2, "multipl_e_javascript": 86.8, "multipl_e_julia": 64.5, "multipl_e_kotlin": 74.8, "multipl_e_lua": 57.8, "multipl_e_perl": 52.5, "multipl_e_php": 79.5, "multipl_e_python": 92.8, "multipl_e_r": 60.2, "multipl_e_ruby": 67.2, "multipl_e_rust": 75.1, "multipl_e_scala": 62.5, "multipl_e_swift": 67.5, "multipl_e_typescript": 86.1, "musr": 33.5, "swe_bench_agent": 42.5, "swe_bench_verified": 53.8, "tau_bench": 48.8, "terminal_bench": 5.7, "wildbench": 72.5}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.55, "display_name": "DeepSeek R1 0528 NVFP4 v2", "fits": {"apple_m3_ultra": 1.7}, "model_id": "nvidia/deepseek-r1-0528-nvfp4-v2", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 54.8, "arena_elo_coding": 1300.0, "arena_elo_math": 1280.0, "arena_elo_overall": 1290.0, "artificial_analysis_quality_index": 73.0, "artificial_analysis_speed_index": 84.0, "bbh": 62.3, "chartqa": 78.5, "docvqa": 86.8, "gpqa_diamond": 67.8, "humaneval": 84.1, "ifeval": 83.0, "math_500": 87.3, "mathvista": 54.5, "mmlu_astronomy": 73.5, "mmlu_biology": 81.5, "mmlu_business_ethics": 74.2, "mmlu_chemistry": 74.5, "mmlu_clinical_knowledge": 80.2, "mmlu_computer_science": 82.5, "mmlu_jurisprudence": 71.5, "mmlu_physics": 78.8, "mmlu_pro": 74.1, "mmlu_professional_accounting": 65.2, "mmlu_professional_law": 70.2, "mmmu": 60.2, "multipl_e": 77.3, "multipl_e_cpp": 76.5, "multipl_e_csharp": 74.2, "multipl_e_go": 73.2, "multipl_e_java": 81.2, "multipl_e_javascript": 79.5, "multipl_e_julia": 54.2, "multipl_e_kotlin": 65.2, "multipl_e_lua": 47.5, "multipl_e_perl": 42.5, "multipl_e_php": 71.5, "multipl_e_python": 84.1, "multipl_e_r": 50.5, "multipl_e_ruby": 57.8, "multipl_e_rust": 67.5, "multipl_e_scala": 52.5, "multipl_e_swift": 57.8, "multipl_e_typescript": 78.8, "musr": 30.4, "swe_bench_verified": 30.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "Gemma 4 31B IT NVFP4", "fits": {"amd_rx_7900_xtx": 43.4, "apple_m1_max": 18.1, "apple_m2_max": 18.1, "apple_m2_ultra": 36.1, "apple_m3_max": 18.1, "apple_m3_ultra": 37.0, "apple_m4": 5.4, "apple_m4_max": 24.7, "apple_m4_pro": 12.3, "nvidia_rtx_3090": 42.3, "nvidia_rtx_4090": 45.5, "nvidia_rtx_5090": 80.9}, "model_id": "nvidia/gemma-4-31b-it-nvfp4", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 5 NVFP4", "fits": {"apple_m3_ultra": 1.5}, "model_id": "nvidia/glm-5-nvfp4", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {"gsm8k": 97.3}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Kimi K2.5 NVFP4", "fits": {}, "model_id": "nvidia/kimi-k2-5-nvfp4", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1160.0, "arena_elo_math": 1150.0, "arena_elo_overall": 1170.0, "artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0, "bbh": 39.2, "bbq": 72.8, "gpqa_diamond": 28.5, "helm_safety": 78.5, "humaneval": 55.5, "ifeval": 62.0, "math_500": 42.0, "mmlu_pro": 40.5, "musr": 12.8, "toxigen": 82.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Llama 3.1 8B Instruct FP8", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nvidia/llama-3-1-8b-instruct-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1160.0, "arena_elo_math": 1150.0, "arena_elo_overall": 1170.0, "artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0, "bbh": 39.2, "bbq": 72.8, "gpqa_diamond": 28.5, "helm_safety": 78.5, "humaneval": 55.5, "ifeval": 62.0, "math_500": 42.0, "mmlu_pro": 40.5, "musr": 12.8, "toxigen": 82.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Llama 3.1 8B Instruct NVFP4", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nvidia/llama-3-1-8b-instruct-nvfp4", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "Llama 3.1 Nemotron Nano 8B v1", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nvidia/llama-3-1-nemotron-nano-8b-v1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.3, "display_name": "Llama 3.1 Nemotron Nano VL 8B V1", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "nvidia/llama-3-1-nemotron-nano-vl-8b-v1", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {"aider_polyglot": 45.2, "alpaca_eval": 40.5, "arena_elo_coding": 1250.0, "arena_elo_math": 1240.0, "arena_elo_overall": 1260.0, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 73.0, "bbh": 58.1, "bbq": 80.1, "gpqa_diamond": 50.2, "gsm8k": 95.0, "helm_safety": 84.2, "humaneval": 80.5, "ifeval": 78.5, "live_code_bench": 28.8, "math_500": 73.2, "mgsm": 91.1, "mmlu_astronomy": 71.5, "mmlu_biology": 80.2, "mmlu_business_ethics": 72.8, "mmlu_chemistry": 72.5, "mmlu_clinical_knowledge": 78.5, "mmlu_computer_science": 80.5, "mmlu_jurisprudence": 69.5, "mmlu_physics": 76.8, "mmlu_pro": 66.5, "mmlu_professional_accounting": 63.8, "mmlu_professional_law": 68.2, "mt_bench": 8.6, "multipl_e": 72.9, "multipl_e_cpp": 72.8, "multipl_e_csharp": 72.5, "multipl_e_go": 68.5, "multipl_e_java": 77.5, "multipl_e_javascript": 75.8, "multipl_e_julia": 52.2, "multipl_e_kotlin": 62.5, "multipl_e_lua": 45.8, "multipl_e_perl": 40.5, "multipl_e_php": 70.2, "multipl_e_python": 80.2, "multipl_e_r": 48.5, "multipl_e_ruby": 55.8, "multipl_e_rust": 60.5, "multipl_e_scala": 50.2, "multipl_e_swift": 55.5, "multipl_e_typescript": 74.8, "musr": 24.5, "swe_bench_verified": 25.8, "toxigen": 88.2, "wildbench": 68.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.88, "display_name": "Llama 3.3 70B Instruct NVFP4", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "nvidia/llama-3-3-70b-instruct-nvfp4", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "Llama 3 3 Nemotron Super 49B v1 5", "fits": {"apple_m1_max": 11.4, "apple_m2_max": 11.4, "apple_m2_ultra": 22.9, "apple_m3_max": 11.4, "apple_m3_ultra": 23.4, "apple_m4_max": 15.6, "apple_m4_pro": 7.8}, "model_id": "nvidia/llama-3-3-nemotron-super-49b-v1-5", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.3, "display_name": "llama nemotron rerank 1B v2", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "nvidia/llama-nemotron-rerank-1b-v2", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "mit b0", "fits": {"amd_rx_7900_xt": 302702.7, "amd_rx_7900_xtx": 363243.2, "amd_rx_9070_xt": 244054.1, "apple_m1_max": 151351.4, "apple_m2_max": 151351.4, "apple_m2_ultra": 302702.7, "apple_m3_max": 151351.4, "apple_m3_ultra": 309891.9, "apple_m4": 45405.4, "apple_m4_max": 206594.6, "apple_m4_pro": 103297.3, "nvidia_rtx_3060_12gb": 136216.2, "nvidia_rtx_3090": 354162.2, "nvidia_rtx_4060_ti_16gb": 108973.0, "nvidia_rtx_4070_ti_super": 254270.3, "nvidia_rtx_4080_super": 278486.5, "nvidia_rtx_4090": 381405.4, "nvidia_rtx_5080": 363243.2, "nvidia_rtx_5090": 678054.1}, "model_id": "nvidia/mit-b0", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "Nemotron Cascade 2 30B A3B", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "nvidia/nemotron-cascade-2-30b-a3b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "Nemotron H 4B Instruct 128K", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "nvidia/nemotron-h-4b-instruct-128k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {"beir": 58.5, "miracl": 55.5, "mteb_classification": 73.5, "mteb_clustering": 53.8, "mteb_overall": 69.5, "mteb_pair_classification": 84.1, "mteb_reranking": 60.4, "mteb_retrieval": 67.2, "mteb_sts": 79.6, "mteb_summarization": 30.4}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "NV Embed v2", "fits": {"amd_rx_7900_xt": 143.6, "amd_rx_7900_xtx": 172.3, "amd_rx_9070_xt": 115.8, "apple_m1_max": 71.8, "apple_m2_max": 71.8, "apple_m2_ultra": 143.6, "apple_m3_max": 71.8, "apple_m3_ultra": 147.0, "apple_m4": 21.5, "apple_m4_max": 98.0, "apple_m4_pro": 49.0, "nvidia_rtx_3060_12gb": 64.6, "nvidia_rtx_3090": 168.0, "nvidia_rtx_4060_ti_16gb": 51.7, "nvidia_rtx_4070_ti_super": 120.6, "nvidia_rtx_4080_super": 132.1, "nvidia_rtx_4090": 180.9, "nvidia_rtx_5080": 172.3, "nvidia_rtx_5090": 321.6}, "model_id": "nvidia/nv-embed-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron 3 Nano 30B A3B Base BF16", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "nvidia/nvidia-nemotron-3-nano-30b-a3b-base-bf16", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron 3 Nano 30B A3B BF16", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "nvidia/nvidia-nemotron-3-nano-30b-a3b-bf16", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron 3 Nano 30B A3B FP8", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "nvidia/nvidia-nemotron-3-nano-30b-a3b-fp8", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron 3 Nano 30B A3B NVFP4", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "nvidia/nvidia-nemotron-3-nano-30b-a3b-nvfp4", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron 3 Super 120B A12B BF16", "fits": {"apple_m2_max": 4.7, "apple_m2_ultra": 9.3, "apple_m3_max": 4.7, "apple_m3_ultra": 9.6, "apple_m4_max": 6.4}, "model_id": "nvidia/nvidia-nemotron-3-super-120b-a12b-bf16", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron 3 Super 120B A12B FP8", "fits": {"apple_m2_max": 4.7, "apple_m2_ultra": 9.3, "apple_m3_max": 4.7, "apple_m3_ultra": 9.6, "apple_m4_max": 6.4}, "model_id": "nvidia/nvidia-nemotron-3-super-120b-a12b-fp8", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron 3 Super 120B A12B NVFP4", "fits": {"apple_m2_max": 4.7, "apple_m2_ultra": 9.3, "apple_m3_max": 4.7, "apple_m3_ultra": 9.6, "apple_m4_max": 6.4}, "model_id": "nvidia/nvidia-nemotron-3-super-120b-a12b-nvfp4", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron Nano 12B v2 VL BF16", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "nvidia/nvidia-nemotron-nano-12b-v2-vl-bf16", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron Nano 12B v2 VL FP8", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "nvidia/nvidia-nemotron-nano-12b-v2-vl-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron Nano 9B v2 Base", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "nvidia/nvidia-nemotron-nano-9b-v2-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron Nano 9B v2 FP8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "nvidia/nvidia-nemotron-nano-9b-v2-fp8", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron Nano 9B v2 Japanese", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "nvidia/nvidia-nemotron-nano-9b-v2-japanese", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron Nano 9B v2", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "nvidia/nvidia-nemotron-nano-9b-v2", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.3, "display_name": "NVIDIA Nemotron Parse v1.1", "fits": {}, "model_id": "nvidia/nvidia-nemotron-parse-v1-1", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {"ai2d": 94.2, "chartqa": 86.0, "docvqa": 92.6, "mathvista": 65.2, "mmmu": 58.7, "ocrbench": 85.2, "realworldqa": 69.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "NVLM D 72B", "fits": {"apple_m1_max": 7.8, "apple_m2_max": 7.8, "apple_m2_ultra": 15.6, "apple_m3_max": 7.8, "apple_m3_ultra": 15.9, "apple_m4_max": 10.6, "apple_m4_pro": 5.3}, "model_id": "nvidia/nvlm-d-72b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "OpenReasoning Nemotron 32B", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "nvidia/openreasoning-nemotron-32b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "parakeet ctc 1.1B", "fits": {"amd_rx_7900_xt": 1018.2, "amd_rx_7900_xtx": 1221.8, "amd_rx_9070_xt": 820.9, "apple_m1_max": 509.1, "apple_m2_max": 509.1, "apple_m2_ultra": 1018.2, "apple_m3_max": 509.1, "apple_m3_ultra": 1042.4, "apple_m4": 152.7, "apple_m4_max": 694.9, "apple_m4_pro": 347.5, "nvidia_rtx_3060_12gb": 458.2, "nvidia_rtx_3090": 1191.3, "nvidia_rtx_4060_ti_16gb": 366.5, "nvidia_rtx_4070_ti_super": 855.3, "nvidia_rtx_4080_super": 936.7, "nvidia_rtx_4090": 1282.9, "nvidia_rtx_5080": 1221.8, "nvidia_rtx_5090": 2280.7}, "model_id": "nvidia/parakeet-ctc-1-1b", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "parakeet tdt 0.6B v2", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "nvidia/parakeet-tdt-0-6b-v2", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "parakeet tdt 0.6B v3", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "nvidia/parakeet-tdt-0-6b-v3", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "personaplex 7B v1", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "nvidia/personaplex-7b-v1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1280.0, "arena_elo_math": 1290.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 70.0, "artificial_analysis_speed_index": 80.0, "bbh": 55.1, "gpqa_diamond": 50.1, "humaneval": 76.5, "ifeval": 77.0, "math_500": 80.5, "mmlu_pro": 64.2, "musr": 24.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.2, "display_name": "Qwen3 14B NVFP4", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "nvidia/qwen3-14b-nvfp4", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1280.0, "artificial_analysis_quality_index": 74.0, "artificial_analysis_speed_index": 88.0, "bbh": 50.5, "gpqa_diamond": 55.2, "humaneval": 80.5, "ifeval": 80.0, "live_code_bench": 62.6, "math_500": 84.2, "mmlu_pro": 68.5, "musr": 21.3}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.15, "display_name": "Qwen3 30B A3B NVFP4", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "nvidia/qwen3-30b-a3b-nvfp4", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 40.0, "alpaca_eval": 38.2, "arena_elo_coding": 1320.0, "arena_elo_math": 1330.0, "arena_elo_overall": 1310.0, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 70.0, "bbh": 62.8, "gpqa_diamond": 62.5, "humaneval": 85.2, "ifeval": 84.0, "live_code_bench": 54.6, "math_500": 88.5, "mmlu_pro": 74.8, "mt_bench": 8.5, "multipl_e_csharp": 78.2, "multipl_e_julia": 59.2, "multipl_e_kotlin": 70.5, "multipl_e_lua": 53.5, "multipl_e_perl": 48.5, "multipl_e_php": 75.8, "multipl_e_r": 55.5, "multipl_e_ruby": 63.5, "multipl_e_scala": 58.2, "multipl_e_swift": 63.8, "musr": 30.1, "wildbench": 65.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.3, "display_name": "Qwen3 32B NVFP4", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "nvidia/qwen3-32b-nvfp4", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 1.2, "display_name": "Qwen3.5 397B A17B NVFP4", "fits": {"apple_m3_ultra": 2.9}, "model_id": "nvidia/qwen3-5-397b-a17b-nvfp4", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "segformer b0 finetuned ade 512 512", "fits": {"amd_rx_7900_xt": 302702.7, "amd_rx_7900_xtx": 363243.2, "amd_rx_9070_xt": 244054.1, "apple_m1_max": 151351.4, "apple_m2_max": 151351.4, "apple_m2_ultra": 302702.7, "apple_m3_max": 151351.4, "apple_m3_ultra": 309891.9, "apple_m4": 45405.4, "apple_m4_max": 206594.6, "apple_m4_pro": 103297.3, "nvidia_rtx_3060_12gb": 136216.2, "nvidia_rtx_3090": 354162.2, "nvidia_rtx_4060_ti_16gb": 108973.0, "nvidia_rtx_4070_ti_super": 254270.3, "nvidia_rtx_4080_super": 278486.5, "nvidia_rtx_4090": 381405.4, "nvidia_rtx_5080": 363243.2, "nvidia_rtx_5090": 678054.1}, "model_id": "nvidia/segformer-b0-finetuned-ade-512-512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "segformer b1 finetuned ade 512 512", "fits": {"amd_rx_7900_xt": 81751.8, "amd_rx_7900_xtx": 98102.2, "amd_rx_9070_xt": 65912.4, "apple_m1_max": 40875.9, "apple_m2_max": 40875.9, "apple_m2_ultra": 81751.8, "apple_m3_max": 40875.9, "apple_m3_ultra": 83693.4, "apple_m4": 12262.8, "apple_m4_max": 55795.6, "apple_m4_pro": 27897.8, "nvidia_rtx_3060_12gb": 36788.3, "nvidia_rtx_3090": 95649.6, "nvidia_rtx_4060_ti_16gb": 29430.7, "nvidia_rtx_4070_ti_super": 68671.5, "nvidia_rtx_4080_super": 75211.7, "nvidia_rtx_4090": 103007.3, "nvidia_rtx_5080": 98102.2, "nvidia_rtx_5090": 183124.1}, "model_id": "nvidia/segformer-b1-finetuned-ade-512-512", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "segformer b5 finetuned ade 640 640", "fits": {"amd_rx_7900_xt": 13223.1, "amd_rx_7900_xtx": 15867.8, "amd_rx_9070_xt": 10661.2, "apple_m1_max": 6611.6, "apple_m2_max": 6611.6, "apple_m2_ultra": 13223.1, "apple_m3_max": 6611.6, "apple_m3_ultra": 13537.2, "apple_m4": 1983.5, "apple_m4_max": 9024.8, "apple_m4_pro": 4512.4, "nvidia_rtx_3060_12gb": 5950.4, "nvidia_rtx_3090": 15471.1, "nvidia_rtx_4060_ti_16gb": 4760.3, "nvidia_rtx_4070_ti_super": 11107.4, "nvidia_rtx_4080_super": 12165.3, "nvidia_rtx_4090": 16661.2, "nvidia_rtx_5080": 15867.8, "nvidia_rtx_5090": 29619.8}, "model_id": "nvidia/segformer-b5-finetuned-ade-640-640", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "segformer b5 finetuned cityscapes 1024 1024", "fits": {"amd_rx_7900_xt": 13223.1, "amd_rx_7900_xtx": 15867.8, "amd_rx_9070_xt": 10661.2, "apple_m1_max": 6611.6, "apple_m2_max": 6611.6, "apple_m2_ultra": 13223.1, "apple_m3_max": 6611.6, "apple_m3_ultra": 13537.2, "apple_m4": 1983.5, "apple_m4_max": 9024.8, "apple_m4_pro": 4512.4, "nvidia_rtx_3060_12gb": 5950.4, "nvidia_rtx_3090": 15471.1, "nvidia_rtx_4060_ti_16gb": 4760.3, "nvidia_rtx_4070_ti_super": 11107.4, "nvidia_rtx_4080_super": 12165.3, "nvidia_rtx_4090": 16661.2, "nvidia_rtx_5080": 15867.8, "nvidia_rtx_5090": 29619.8}, "model_id": "nvidia/segformer-b5-finetuned-cityscapes-1024-1024", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "speakerverification en titanet large", "fits": {"amd_rx_7900_xt": 48695.7, "amd_rx_7900_xtx": 58434.8, "amd_rx_9070_xt": 39260.9, "apple_m1_max": 24347.8, "apple_m2_max": 24347.8, "apple_m2_ultra": 48695.7, "apple_m3_max": 24347.8, "apple_m3_ultra": 49852.2, "apple_m4": 7304.3, "apple_m4_max": 33234.8, "apple_m4_pro": 16617.4, "nvidia_rtx_3060_12gb": 21913.0, "nvidia_rtx_3090": 56973.9, "nvidia_rtx_4060_ti_16gb": 17530.4, "nvidia_rtx_4070_ti_super": 40904.3, "nvidia_rtx_4080_super": 44800.0, "nvidia_rtx_4090": 61356.5, "nvidia_rtx_5080": 58434.8, "nvidia_rtx_5090": 109078.3}, "model_id": "nvidia/speakerverification-en-titanet-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "NVIDIA", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 0, "cost_input": null, "display_name": "chatgpt-image-latest", "fits": {}, "model_id": "openai/chatgpt-image-latest", "model_subtypes": [], "model_type": "image-generation", "open_weights": false, "provider": "OpenAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 1.5, "display_name": "Codex Mini", "fits": {}, "model_id": "openai/codex-mini-latest", "model_subtypes": [], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": null}, {"benchmark_scores": {"alpaca_eval": 25.4, "arena_elo_coding": 1100.0, "arena_elo_math": 1080.0, "arena_elo_overall": 1110.0, "artificial_analysis_quality_index": 60.0, "artificial_analysis_speed_index": 92.0, "gpqa_diamond": 28.3, "gsm8k": 80.0, "humaneval": 48.1, "ifeval": 60.5, "math_500": 32.8, "mgsm": 56.3, "mmlu_pro": 40.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 16385, "cost_input": 0.5, "display_name": "GPT-3.5-turbo", "fits": {}, "model_id": "openai/gpt-3-5-turbo", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 91.5, "aider_polyglot": 32.4, "alpaca_eval": 52.1, "arena_elo_coding": 1310.0, "arena_elo_hard_prompts": 1402.3, "arena_elo_math": 1290.0, "arena_elo_overall": 1300.0, "arena_elo_style_control": 1382.1, "arena_elo_vision": 1202.1, "artificial_analysis_quality_index": 74.0, "artificial_analysis_speed_index": 89.0, "bbq": 86.5, "chartqa": 88.5, "docvqa": 94.1, "gpqa_diamond": 60.1, "helm_safety": 90.5, "humaneval": 86.5, "ifeval": 85.0, "live_code_bench": 48.3, "math_500": 82.3, "mathvista": 68.2, "mmlu_pro": 74.2, "mmmu": 72.5, "mt_bench": 9.3, "multipl_e_csharp": 76.5, "multipl_e_julia": 56.2, "multipl_e_kotlin": 68.2, "multipl_e_lua": 50.8, "multipl_e_perl": 46.5, "multipl_e_php": 74.2, "multipl_e_r": 52.8, "multipl_e_ruby": 62.5, "multipl_e_scala": 56.8, "multipl_e_swift": 61.5, "swe_bench_verified": 23.6, "toxigen": 93.5, "wildbench": 80.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1047576, "cost_input": 0.4, "display_name": "GPT-4.1 mini", "fits": {}, "model_id": "openai/gpt-4-1-mini", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 95.2, "aider_polyglot": 8.9, "alpaca_eval": 52.1, "arena_elo_coding": 1260.0, "arena_elo_hard_prompts": 1332.8, "arena_elo_math": 1240.0, "arena_elo_overall": 1250.0, "arena_elo_style_control": 1321.4, "arena_elo_vision": 1088.9, "artificial_analysis_quality_index": 65.0, "artificial_analysis_speed_index": 94.0, "bbq": 86.5, "chartqa": 88.5, "docvqa": 94.1, "gpqa_diamond": 48.3, "helm_safety": 90.5, "humaneval": 78.5, "ifeval": 78.0, "live_code_bench": 48.3, "math_500": 70.5, "mathvista": 68.2, "mmlu_pro": 64.1, "mmmu": 72.5, "mt_bench": 9.3, "multipl_e_csharp": 76.5, "multipl_e_julia": 56.2, "multipl_e_kotlin": 68.2, "multipl_e_lua": 50.8, "multipl_e_perl": 46.5, "multipl_e_php": 74.2, "multipl_e_r": 52.8, "multipl_e_ruby": 62.5, "multipl_e_scala": 56.8, "multipl_e_swift": 61.5, "swe_bench_verified": 54.6, "toxigen": 93.5, "wildbench": 80.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1047576, "cost_input": 0.1, "display_name": "GPT-4.1 nano", "fits": {}, "model_id": "openai/gpt-4-1-nano", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 95.2, "aider_polyglot": 68.5, "alpaca_eval": 52.1, "arena_elo_coding": 1360.0, "arena_elo_hard_prompts": 1430.8, "arena_elo_math": 1320.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1413.0, "arena_elo_vision": 1213.5, "artificial_analysis_quality_index": 84.0, "artificial_analysis_speed_index": 76.0, "bbq": 86.5, "chartqa": 88.5, "docvqa": 94.1, "gpqa_diamond": 69.5, "helm_safety": 90.5, "humaneval": 90.2, "ifeval": 89.5, "live_code_bench": 48.3, "math_500": 89.4, "mathvista": 68.2, "medqa": 89.7, "mmlu_astronomy": 79.5, "mmlu_biology": 87.5, "mmlu_business_ethics": 81.5, "mmlu_chemistry": 81.8, "mmlu_clinical_knowledge": 86.2, "mmlu_computer_science": 88.2, "mmlu_jurisprudence": 79.8, "mmlu_physics": 84.2, "mmlu_pro": 80.1, "mmlu_professional_accounting": 71.8, "mmlu_professional_law": 77.5, "mmmu": 72.5, "mt_bench": 9.3, "multipl_e": 85.7, "multipl_e_cpp": 85.1, "multipl_e_csharp": 85.2, "multipl_e_go": 82.1, "multipl_e_java": 88.5, "multipl_e_javascript": 88.2, "multipl_e_julia": 65.8, "multipl_e_kotlin": 78.5, "multipl_e_lua": 61.5, "multipl_e_perl": 58.2, "multipl_e_php": 82.8, "multipl_e_python": 92.5, "multipl_e_r": 63.5, "multipl_e_ruby": 73.5, "multipl_e_rust": 76.2, "multipl_e_scala": 67.2, "multipl_e_swift": 71.8, "multipl_e_typescript": 87.1, "swe_bench_agent": 48.5, "swe_bench_verified": 54.6, "tau_bench": 52.8, "terminal_bench": 42.1, "toxigen": 93.5, "wildbench": 80.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1047576, "cost_input": 2.0, "display_name": "GPT-4.1", "fits": {}, "model_id": "openai/gpt-4-1", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 89.4, "alpaca_eval": 55.0, "arena_elo_coding": 1270.0, "arena_elo_hard_prompts": 1315.4, "arena_elo_math": 1250.0, "arena_elo_overall": 1260.0, "arena_elo_style_control": 1323.4, "arena_elo_vision": 1112.5, "artificial_analysis_quality_index": 79.0, "artificial_analysis_speed_index": 65.0, "bbq": 85.5, "chartqa": 87.2, "docvqa": 87.2, "flores_en_de": 67.3, "flores_en_es": 71.2, "flores_en_ja": 61.4, "flores_en_zh": 59.7, "gpqa_diamond": 49.3, "gsm8k": 93.0, "helm_safety": 89.5, "humaneval": 87.8, "ifeval": 82.0, "live_code_bench": 29.1, "math_500": 72.2, "mathvista": 50.5, "mgsm": 88.5, "mmlu_pro": 67.4, "mmmu": 59.4, "multipl_e_csharp": 80.5, "multipl_e_julia": 61.8, "multipl_e_kotlin": 73.8, "multipl_e_lua": 56.8, "multipl_e_perl": 52.5, "multipl_e_php": 78.2, "multipl_e_r": 58.2, "multipl_e_ruby": 68.5, "multipl_e_scala": 62.8, "multipl_e_swift": 66.5, "toxigen": 93.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 10.0, "display_name": "GPT-4 Turbo", "fits": {}, "model_id": "openai/gpt-4-turbo", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 95.2, "aider_polyglot": 52.4, "alpaca_eval": 52.1, "arena_elo_coding": 1240.0, "arena_elo_math": 1220.0, "arena_elo_overall": 1230.0, "bbq": 86.5, "chartqa": 88.5, "docvqa": 94.1, "flores_en_de": 67.3, "flores_en_es": 71.2, "flores_en_ja": 61.4, "flores_en_zh": 59.7, "gpqa_diamond": 39.7, "gsm8k": 92.0, "helm_safety": 90.5, "humaneval": 67.0, "ifeval": 76.5, "live_code_bench": 48.3, "math_500": 42.5, "mathvista": 68.2, "mgsm": 74.5, "mmlu_pro": 56.2, "mmmu": 72.5, "mt_bench": 9.3, "multipl_e_csharp": 76.5, "multipl_e_julia": 56.2, "multipl_e_kotlin": 68.2, "multipl_e_lua": 50.8, "multipl_e_perl": 46.5, "multipl_e_php": 74.2, "multipl_e_r": 52.8, "multipl_e_ruby": 62.5, "multipl_e_scala": 56.8, "multipl_e_swift": 61.5, "swe_bench_verified": 54.6, "toxigen": 93.5, "wildbench": 80.5}, "capability_tiers": {"coding": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 30.0, "display_name": "GPT-4", "fits": {}, "model_id": "openai/gpt-4", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 94.2, "aider_polyglot": 58.2, "alpaca_eval": 48.5, "arena_elo_coding": 1300.0, "arena_elo_hard_prompts": 1336.8, "arena_elo_math": 1270.0, "arena_elo_overall": 1290.0, "arena_elo_style_control": 1345.1, "arena_elo_vision": 1310.0, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 78.0, "bbq": 85.8, "chartqa": 85.2, "docvqa": 92.8, "finbench": 68.5, "flores_en_de": 67.3, "flores_en_es": 71.2, "flores_en_ja": 61.4, "flores_en_zh": 59.7, "gpqa_diamond": 53.6, "gsm8k": 95.0, "helm_safety": 89.2, "humaneval": 90.2, "ifeval": 84.3, "legalbench": 72.1, "live_code_bench": 31.7, "math_500": 76.6, "mathvista": 63.8, "medqa": 78.5, "mgsm": 90.5, "mmlu_astronomy": 78.2, "mmlu_biology": 86.8, "mmlu_business_ethics": 80.2, "mmlu_chemistry": 80.2, "mmlu_clinical_knowledge": 85.5, "mmlu_computer_science": 87.5, "mmlu_jurisprudence": 78.5, "mmlu_physics": 83.5, "mmlu_pro": 72.6, "mmlu_professional_accounting": 70.5, "mmlu_professional_law": 76.8, "mmmu": 69.1, "mt_bench": 9.1, "multipl_e": 83.0, "multipl_e_cpp": 82.5, "multipl_e_csharp": 82.1, "multipl_e_go": 79.1, "multipl_e_java": 86.5, "multipl_e_javascript": 85.8, "multipl_e_julia": 62.8, "multipl_e_kotlin": 75.8, "multipl_e_lua": 58.5, "multipl_e_perl": 55.2, "multipl_e_php": 80.5, "multipl_e_python": 90.2, "multipl_e_r": 60.5, "multipl_e_ruby": 70.2, "multipl_e_rust": 72.8, "multipl_e_scala": 64.5, "multipl_e_swift": 68.2, "multipl_e_typescript": 84.2, "ocrbench": 73.6, "realworldqa": 75.4, "swe_bench_agent": 32.1, "swe_bench_verified": 38.4, "tau_bench": 42.5, "terminal_bench": 35.2, "toxigen": 92.8, "wildbench": 75.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 5.0, "display_name": "GPT-4o (2024-05-13)", "fits": {}, "model_id": "openai/gpt-4o-2024-05-13", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 94.2, "aider_polyglot": 58.2, "alpaca_eval": 48.5, "arena_elo_coding": 1300.0, "arena_elo_hard_prompts": 1326.3, "arena_elo_math": 1270.0, "arena_elo_overall": 1290.0, "arena_elo_style_control": 1334.3, "arena_elo_vision": 1310.0, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 78.0, "bbq": 85.8, "chartqa": 85.2, "docvqa": 92.8, "finbench": 68.5, "flores_en_de": 67.3, "flores_en_es": 71.2, "flores_en_ja": 61.4, "flores_en_zh": 59.7, "gpqa_diamond": 53.6, "gsm8k": 95.0, "helm_safety": 89.2, "humaneval": 90.2, "ifeval": 84.3, "legalbench": 72.1, "live_code_bench": 31.7, "math_500": 76.6, "mathvista": 63.8, "medqa": 78.5, "mgsm": 90.5, "mmlu_astronomy": 78.2, "mmlu_biology": 86.8, "mmlu_business_ethics": 80.2, "mmlu_chemistry": 80.2, "mmlu_clinical_knowledge": 85.5, "mmlu_computer_science": 87.5, "mmlu_jurisprudence": 78.5, "mmlu_physics": 83.5, "mmlu_pro": 72.6, "mmlu_professional_accounting": 70.5, "mmlu_professional_law": 76.8, "mmmu": 69.1, "mt_bench": 9.1, "multipl_e": 83.0, "multipl_e_cpp": 82.5, "multipl_e_csharp": 82.1, "multipl_e_go": 79.1, "multipl_e_java": 86.5, "multipl_e_javascript": 85.8, "multipl_e_julia": 62.8, "multipl_e_kotlin": 75.8, "multipl_e_lua": 58.5, "multipl_e_perl": 55.2, "multipl_e_php": 80.5, "multipl_e_python": 90.2, "multipl_e_r": 60.5, "multipl_e_ruby": 70.2, "multipl_e_rust": 72.8, "multipl_e_scala": 64.5, "multipl_e_swift": 68.2, "multipl_e_typescript": 84.2, "ocrbench": 73.6, "realworldqa": 75.4, "swe_bench_agent": 32.1, "swe_bench_verified": 38.4, "tau_bench": 42.5, "terminal_bench": 35.2, "toxigen": 92.8, "wildbench": 75.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 2.5, "display_name": "GPT-4o (2024-08-06)", "fits": {}, "model_id": "openai/gpt-4o-2024-08-06", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 94.2, "aider_polyglot": 58.2, "alpaca_eval": 48.5, "arena_elo_coding": 1300.0, "arena_elo_math": 1270.0, "arena_elo_overall": 1290.0, "arena_elo_vision": 1310.0, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 78.0, "bbq": 85.8, "chartqa": 85.2, "docvqa": 92.8, "finbench": 68.5, "flores_en_de": 67.3, "flores_en_es": 71.2, "flores_en_ja": 61.4, "flores_en_zh": 59.7, "gpqa_diamond": 53.6, "gsm8k": 95.0, "helm_safety": 89.2, "humaneval": 90.2, "ifeval": 84.3, "legalbench": 72.1, "live_code_bench": 31.7, "math_500": 76.6, "mathvista": 63.8, "medqa": 78.5, "mgsm": 90.5, "mmlu_astronomy": 78.2, "mmlu_biology": 86.8, "mmlu_business_ethics": 80.2, "mmlu_chemistry": 80.2, "mmlu_clinical_knowledge": 85.5, "mmlu_computer_science": 87.5, "mmlu_jurisprudence": 78.5, "mmlu_physics": 83.5, "mmlu_pro": 72.6, "mmlu_professional_accounting": 70.5, "mmlu_professional_law": 76.8, "mmmu": 69.1, "mt_bench": 9.1, "multipl_e": 83.0, "multipl_e_cpp": 82.5, "multipl_e_csharp": 82.1, "multipl_e_go": 79.1, "multipl_e_java": 86.5, "multipl_e_javascript": 85.8, "multipl_e_julia": 62.8, "multipl_e_kotlin": 75.8, "multipl_e_lua": 58.5, "multipl_e_perl": 55.2, "multipl_e_php": 80.5, "multipl_e_python": 90.2, "multipl_e_r": 60.5, "multipl_e_ruby": 70.2, "multipl_e_rust": 72.8, "multipl_e_scala": 64.5, "multipl_e_swift": 68.2, "multipl_e_typescript": 84.2, "ocrbench": 73.6, "realworldqa": 75.4, "swe_bench_agent": 32.1, "swe_bench_verified": 38.4, "tau_bench": 42.5, "terminal_bench": 35.2, "toxigen": 92.8, "wildbench": 75.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 2.5, "display_name": "GPT-4o (2024-11-20)", "fits": {}, "model_id": "openai/gpt-4o-2024-11-20", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 58.2, "alpaca_eval": 48.5, "arena_elo_coding": 1260.0, "arena_elo_hard_prompts": 1311.1, "arena_elo_math": 1240.0, "arena_elo_overall": 1250.0, "arena_elo_style_control": 1317.2, "arena_elo_vision": 1097.8, "artificial_analysis_quality_index": 72.0, "artificial_analysis_speed_index": 90.0, "bbq": 85.8, "chartqa": 85.2, "docvqa": 92.8, "finbench": 68.5, "flores_en_de": 67.3, "flores_en_es": 71.2, "flores_en_ja": 61.4, "flores_en_zh": 59.7, "gpqa_diamond": 46.1, "gsm8k": 92.0, "helm_safety": 89.2, "humaneval": 87.2, "ifeval": 80.5, "legalbench": 72.1, "live_code_bench": 23.4, "math_500": 70.2, "mathvista": 63.8, "medqa": 78.5, "mgsm": 87.0, "mmlu_astronomy": 78.2, "mmlu_biology": 86.8, "mmlu_business_ethics": 80.2, "mmlu_chemistry": 80.2, "mmlu_clinical_knowledge": 85.5, "mmlu_computer_science": 87.5, "mmlu_jurisprudence": 78.5, "mmlu_physics": 83.5, "mmlu_pro": 66.5, "mmlu_professional_accounting": 70.5, "mmlu_professional_law": 76.8, "mmmu": 69.1, "mt_bench": 9.1, "multipl_e": 83.0, "multipl_e_cpp": 82.5, "multipl_e_csharp": 72.5, "multipl_e_go": 79.1, "multipl_e_java": 86.5, "multipl_e_javascript": 85.8, "multipl_e_julia": 53.5, "multipl_e_kotlin": 65.2, "multipl_e_lua": 48.5, "multipl_e_perl": 44.2, "multipl_e_php": 70.2, "multipl_e_python": 90.2, "multipl_e_r": 50.2, "multipl_e_ruby": 58.5, "multipl_e_rust": 72.8, "multipl_e_scala": 52.8, "multipl_e_swift": 58.5, "multipl_e_typescript": 84.2, "swe_bench_agent": 32.1, "swe_bench_verified": 38.5, "tau_bench": 42.5, "terminal_bench": 35.2, "toxigen": 92.8, "wildbench": 75.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 0.15, "display_name": "GPT-4o mini", "fits": {}, "model_id": "openai/gpt-4o-mini", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 94.2, "aider_polyglot": 58.2, "alpaca_eval": 48.5, "arena_elo_coding": 1300.0, "arena_elo_hard_prompts": 1456.0, "arena_elo_math": 1270.0, "arena_elo_overall": 1290.0, "arena_elo_style_control": 1443.0, "arena_elo_vision": 1310.0, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 78.0, "bbq": 85.8, "chartqa": 85.2, "docvqa": 92.8, "finbench": 68.5, "flores_en_de": 67.3, "flores_en_es": 71.2, "flores_en_ja": 61.4, "flores_en_zh": 59.7, "gpqa_diamond": 53.6, "gsm8k": 95.0, "helm_safety": 89.2, "humaneval": 90.2, "ifeval": 84.3, "legalbench": 72.1, "live_code_bench": 31.7, "math_500": 76.6, "mathvista": 63.8, "medqa": 78.5, "mgsm": 90.5, "mmlu_astronomy": 78.2, "mmlu_biology": 86.8, "mmlu_business_ethics": 80.2, "mmlu_chemistry": 80.2, "mmlu_clinical_knowledge": 85.5, "mmlu_computer_science": 87.5, "mmlu_jurisprudence": 78.5, "mmlu_physics": 83.5, "mmlu_pro": 72.6, "mmlu_professional_accounting": 70.5, "mmlu_professional_law": 76.8, "mmmu": 69.1, "mt_bench": 9.1, "multipl_e": 83.0, "multipl_e_cpp": 82.5, "multipl_e_csharp": 82.1, "multipl_e_go": 79.1, "multipl_e_java": 86.5, "multipl_e_javascript": 85.8, "multipl_e_julia": 62.8, "multipl_e_kotlin": 75.8, "multipl_e_lua": 58.5, "multipl_e_perl": 55.2, "multipl_e_php": 80.5, "multipl_e_python": 90.2, "multipl_e_r": 60.5, "multipl_e_ruby": 70.2, "multipl_e_rust": 72.8, "multipl_e_scala": 64.5, "multipl_e_swift": 68.2, "multipl_e_typescript": 84.2, "ocrbench": 73.6, "realworldqa": 75.4, "swe_bench_agent": 32.1, "swe_bench_verified": 38.4, "tau_bench": 42.5, "terminal_bench": 35.2, "toxigen": 92.8, "wildbench": 75.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 2.5, "display_name": "GPT-4o", "fits": {}, "model_id": "openai/gpt-4o", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1410.0, "arena_elo_math": 1390.0, "arena_elo_overall": 1400.0, "artificial_analysis_quality_index": 88.0, "artificial_analysis_speed_index": 70.0, "gpqa_diamond": 79.1, "humaneval": 93.8, "ifeval": 92.5, "live_code_bench": 86.8, "math_500": 96.8, "mmlu_pro": 86.0, "multipl_e_csharp": 89.2, "multipl_e_julia": 71.8, "multipl_e_kotlin": 83.5, "multipl_e_lua": 66.8, "multipl_e_perl": 63.5, "multipl_e_php": 86.5, "multipl_e_r": 69.5, "multipl_e_ruby": 79.8, "multipl_e_scala": 73.8, "multipl_e_swift": 77.8, "swe_bench_agent": 55.2, "swe_bench_verified": 71.5, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 52.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 128000, "cost_input": 1.25, "display_name": "GPT-5.1 Chat", "fits": {}, "model_id": "openai/gpt-5-1-chat-latest", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1410.0, "arena_elo_math": 1390.0, "arena_elo_overall": 1400.0, "artificial_analysis_quality_index": 88.0, "artificial_analysis_speed_index": 70.0, "gpqa_diamond": 79.1, "humaneval": 93.8, "ifeval": 92.5, "live_code_bench": 84.9, "math_500": 96.8, "mmlu_pro": 86.0, "multipl_e_csharp": 89.2, "multipl_e_julia": 71.8, "multipl_e_kotlin": 83.5, "multipl_e_lua": 66.8, "multipl_e_perl": 63.5, "multipl_e_php": 86.5, "multipl_e_r": 69.5, "multipl_e_ruby": 79.8, "multipl_e_scala": 73.8, "multipl_e_swift": 77.8, "swe_bench_agent": 55.2, "swe_bench_verified": 71.5, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 52.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.25, "display_name": "GPT-5.1 Codex Max", "fits": {}, "model_id": "openai/gpt-5-1-codex-max", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1410.0, "arena_elo_math": 1390.0, "arena_elo_overall": 1400.0, "artificial_analysis_quality_index": 88.0, "artificial_analysis_speed_index": 70.0, "gpqa_diamond": 79.1, "humaneval": 93.8, "ifeval": 92.5, "live_code_bench": 83.6, "math_500": 96.8, "mmlu_pro": 86.0, "multipl_e_csharp": 89.2, "multipl_e_julia": 71.8, "multipl_e_kotlin": 83.5, "multipl_e_lua": 66.8, "multipl_e_perl": 63.5, "multipl_e_php": 86.5, "multipl_e_r": 69.5, "multipl_e_ruby": 79.8, "multipl_e_scala": 73.8, "multipl_e_swift": 77.8, "swe_bench_agent": 55.2, "swe_bench_verified": 71.5, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 52.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 0.25, "display_name": "GPT-5.1 Codex mini", "fits": {}, "model_id": "openai/gpt-5-1-codex-mini", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1410.0, "arena_elo_math": 1390.0, "arena_elo_overall": 1400.0, "artificial_analysis_quality_index": 88.0, "artificial_analysis_speed_index": 70.0, "gpqa_diamond": 79.1, "humaneval": 93.8, "ifeval": 92.5, "live_code_bench": 84.9, "math_500": 96.8, "mmlu_pro": 86.0, "multipl_e_csharp": 89.2, "multipl_e_julia": 71.8, "multipl_e_kotlin": 83.5, "multipl_e_lua": 66.8, "multipl_e_perl": 63.5, "multipl_e_php": 86.5, "multipl_e_r": 69.5, "multipl_e_ruby": 79.8, "multipl_e_scala": 73.8, "multipl_e_swift": 77.8, "swe_bench_agent": 55.2, "swe_bench_verified": 71.5, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 52.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.25, "display_name": "GPT-5.1 Codex", "fits": {}, "model_id": "openai/gpt-5-1-codex", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1410.0, "arena_elo_hard_prompts": 1474.6, "arena_elo_math": 1390.0, "arena_elo_overall": 1400.0, "arena_elo_style_control": 1454.6, "arena_elo_vision": 1249.1, "artificial_analysis_quality_index": 88.0, "artificial_analysis_speed_index": 70.0, "gpqa_diamond": 79.1, "humaneval": 93.8, "ifeval": 92.5, "live_code_bench": 86.8, "math_500": 96.8, "medqa": 96.4, "mmlu_pro": 86.0, "multipl_e_csharp": 89.2, "multipl_e_julia": 71.8, "multipl_e_kotlin": 83.5, "multipl_e_lua": 66.8, "multipl_e_perl": 63.5, "multipl_e_php": 86.5, "multipl_e_r": 69.5, "multipl_e_ruby": 79.8, "multipl_e_scala": 73.8, "multipl_e_swift": 77.8, "swe_bench_agent": 55.2, "swe_bench_verified": 71.5, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 52.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.25, "display_name": "GPT-5.1", "fits": {}, "model_id": "openai/gpt-5-1", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 64.0}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 128000, "cost_input": 1.75, "display_name": "GPT-5.2 Chat", "fits": {}, "model_id": "openai/gpt-5-2-chat-latest", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 64.0}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.75, "display_name": "GPT-5.2 Codex", "fits": {}, "model_id": "openai/gpt-5-2-codex", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 88.9, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 21.0, "display_name": "GPT-5.2 Pro", "fits": {}, "model_id": "openai/gpt-5-2-pro", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_hard_prompts": 1495.4, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "arena_elo_style_control": 1477.0, "arena_elo_vision": 1277.9, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "medqa": 95.8, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 64.0}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.75, "display_name": "GPT-5.2", "fits": {}, "model_id": "openai/gpt-5-2", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 77.3}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 128000, "cost_input": 1.75, "display_name": "GPT-5.3 Chat (latest)", "fits": {}, "model_id": "openai/gpt-5-3-chat-latest", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_hard_prompts": 1482.5, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "arena_elo_style_control": 1456.4, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 77.3}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 128000, "cost_input": 1.75, "display_name": "GPT-5.3 Codex Spark", "fits": {}, "model_id": "openai/gpt-5-3-codex-spark", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 77.3}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.75, "display_name": "GPT-5.3 Codex", "fits": {}, "model_id": "openai/gpt-5-3-codex", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_hard_prompts": 1483.9, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "arena_elo_style_control": 1457.8, "arena_elo_vision": 1241.2, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 0.75, "display_name": "GPT-5.4 mini", "fits": {}, "model_id": "openai/gpt-5-4-mini", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_hard_prompts": 1417.2, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "arena_elo_style_control": 1402.7, "arena_elo_vision": 1191.8, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 0.2, "display_name": "GPT-5.4 nano", "fits": {}, "model_id": "openai/gpt-5-4-nano", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 1050000, "cost_input": 30.0, "display_name": "GPT-5.4 Pro", "fits": {}, "model_id": "openai/gpt-5-4-pro", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_hard_prompts": 1507.2, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "arena_elo_style_control": 1483.6, "gpqa_diamond": 92.8, "graphwalks_bfs_256k_1m": 21.4, "hle": 39.8, "hle_tools": 52.1, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "medqa": 93.0, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "osworld": 75.0, "swe_bench_agent": 55.2, "swe_bench_pro": 57.7, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 75.1, "usamo_2026": 95.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 1050000, "cost_input": 2.5, "display_name": "GPT-5.4", "fits": {}, "model_id": "openai/gpt-5-4", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 77.3}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1"}, "context_window": 400000, "cost_input": 1.25, "display_name": "GPT-5 Chat (latest)", "fits": {}, "model_id": "openai/gpt-5-chat-latest", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.0, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.25, "display_name": "GPT-5-Codex", "fits": {}, "model_id": "openai/gpt-5-codex", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1350.0, "arena_elo_hard_prompts": 1401.8, "arena_elo_math": 1330.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1389.5, "arena_elo_vision": 1181.9, "gpqa_diamond": 68.2, "humaneval": 88.5, "ifeval": 88.0, "live_code_bench": 83.8, "math_500": 89.1, "mmlu_pro": 79.5, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 62.8, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 0.25, "display_name": "GPT-5 Mini", "fits": {}, "model_id": "openai/gpt-5-mini", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1300.0, "arena_elo_hard_prompts": 1353.3, "arena_elo_math": 1280.0, "arena_elo_overall": 1290.0, "arena_elo_style_control": 1336.4, "arena_elo_vision": 1145.6, "gpqa_diamond": 55.1, "humaneval": 82.0, "ifeval": 82.5, "live_code_bench": 47.0, "math_500": 78.2, "mmlu_pro": 70.3, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 62.8, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 0.05, "display_name": "GPT-5 Nano", "fits": {}, "model_id": "openai/gpt-5-nano", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "live_code_bench": 84.6, "math_500": 95.5, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 15.0, "display_name": "GPT-5 Pro", "fits": {}, "model_id": "openai/gpt-5-pro", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.1, "arena_elo_coding": 1400.0, "arena_elo_hard_prompts": 1447.9, "arena_elo_math": 1380.0, "arena_elo_overall": 1390.0, "arena_elo_style_control": 1433.4, "arena_elo_vision": 1225.2, "gpqa_diamond": 77.2, "humaneval": 92.1, "ifeval": 91.8, "legalbench": 86.0, "live_code_bench": 84.6, "math_500": 95.5, "medqa": 93.0, "mmlu_pro": 84.8, "multipl_e_csharp": 88.5, "multipl_e_julia": 70.5, "multipl_e_kotlin": 82.8, "multipl_e_lua": 65.5, "multipl_e_perl": 62.8, "multipl_e_php": 85.2, "multipl_e_r": 68.2, "multipl_e_ruby": 78.2, "multipl_e_scala": 72.5, "multipl_e_swift": 76.5, "swe_bench_agent": 55.2, "swe_bench_verified": 69.3, "tau_bench": 58.5, "terminal_bench": 48.8, "terminal_bench_2": 77.3}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 400000, "cost_input": 1.25, "display_name": "GPT-5", "fits": {}, "model_id": "openai/gpt-5", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 0, "cost_input": null, "display_name": "gpt-image-1.5", "fits": {}, "model_id": "openai/gpt-image-1-5", "model_subtypes": [], "model_type": "image-generation", "open_weights": false, "provider": "OpenAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 0, "cost_input": null, "display_name": "gpt-image-1-mini", "fits": {}, "model_id": "openai/gpt-image-1-mini", "model_subtypes": [], "model_type": "image-generation", "open_weights": false, "provider": "OpenAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 0, "cost_input": null, "display_name": "gpt-image-1", "fits": {}, "model_id": "openai/gpt-image-1", "model_subtypes": [], "model_type": "image-generation", "open_weights": false, "provider": "OpenAI", "scores_as_of": null}, {"benchmark_scores": {"aider_polyglot": 32.9, "arena_elo_coding": 1310.0, "arena_elo_hard_prompts": 1360.5, "arena_elo_math": 1330.0, "arena_elo_overall": 1300.0, "arena_elo_style_control": 1336.6, "gpqa_diamond": 60.0, "gsm8k": 97.1, "humaneval": 85.0, "ifeval": 82.0, "math_500": 90.0, "mgsm": 89.3, "mmlu_pro": 72.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1"}, "context_window": 128000, "cost_input": 1.1, "display_name": "o1-mini", "fits": {}, "model_id": "openai/o1-mini", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 61.7, "arena_elo_coding": 1320.0, "arena_elo_hard_prompts": 1396.1, "arena_elo_math": 1360.0, "arena_elo_overall": 1330.0, "arena_elo_style_control": 1387.6, "gpqa_diamond": 73.3, "humaneval": 88.5, "ifeval": 86.0, "live_code_bench": 67.9, "math_500": 92.5, "mgsm": 90.8, "mmlu_pro": 78.0, "swe_bench_verified": 41.3}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1"}, "context_window": 128000, "cost_input": 15.0, "display_name": "o1-preview", "fits": {}, "model_id": "openai/o1-preview", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1360.0, "arena_elo_math": 1400.0, "arena_elo_overall": 1370.0, "gpqa_diamond": 80.5, "gsm8k": 97.1, "humaneval": 90.2, "ifeval": 88.5, "math_500": 95.6, "mgsm": 89.3, "mmlu_pro": 82.0}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 150.0, "display_name": "o1-pro", "fits": {}, "model_id": "openai/o1-pro", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 61.7, "arena_elo_coding": 1320.0, "arena_elo_hard_prompts": 1417.4, "arena_elo_math": 1360.0, "arena_elo_overall": 1330.0, "arena_elo_style_control": 1401.5, "arena_elo_vision": 1192.9, "gpqa_diamond": 73.3, "gsm8k": 97.1, "humaneval": 88.5, "ifeval": 86.0, "live_code_bench": 67.9, "math_500": 92.5, "medqa": 96.5, "mgsm": 89.3, "mmlu_pro": 78.0, "swe_bench_verified": 41.3}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 15.0, "display_name": "o1", "fits": {}, "model_id": "openai/o1", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 80.5, "aime_2025": 91.5, "arena_elo_coding": 1410.0, "arena_elo_math": 1430.0, "arena_elo_overall": 1400.0, "gpqa_diamond": 87.7, "humaneval": 92.8, "ifeval": 91.0, "live_code_bench": 58.3, "math_500": 96.7, "mmlu_pro": 84.1, "swe_bench_agent": 62.8, "swe_bench_verified": 71.7, "tau_bench": 65.5, "terminal_bench": 58.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 10.0, "display_name": "o3-deep-research", "fits": {}, "model_id": "openai/o3-deep-research", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 80.5, "aime_2025": 83.3, "arena_elo_coding": 1350.0, "arena_elo_hard_prompts": 1401.1, "arena_elo_math": 1370.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1363.2, "gpqa_diamond": 79.7, "humaneval": 88.9, "ifeval": 86.5, "live_code_bench": 71.7, "math_500": 94.8, "medqa": 91.4, "mgsm": 92.0, "mmlu_pro": 79.0, "multipl_e_csharp": 82.5, "multipl_e_julia": 64.2, "multipl_e_kotlin": 75.2, "multipl_e_lua": 58.5, "multipl_e_perl": 54.2, "multipl_e_php": 80.2, "multipl_e_r": 60.5, "multipl_e_ruby": 68.5, "multipl_e_scala": 63.2, "multipl_e_swift": 68.2, "swe_bench_agent": 62.8, "swe_bench_verified": 71.2, "tau_bench": 65.5, "terminal_bench": 58.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 1.1, "display_name": "o3-mini", "fits": {}, "model_id": "openai/o3-mini", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 80.5, "aime_2025": 94.0, "arena_elo_coding": 1420.0, "arena_elo_math": 1440.0, "arena_elo_overall": 1410.0, "gpqa_diamond": 89.0, "humaneval": 93.5, "ifeval": 92.0, "math_500": 97.3, "mmlu_pro": 85.5, "swe_bench_agent": 62.8, "swe_bench_verified": 73.2, "tau_bench": 65.5, "terminal_bench": 58.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 20.0, "display_name": "o3-pro", "fits": {}, "model_id": "openai/o3-pro", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 80.5, "aime_2025": 91.5, "arena_elo_coding": 1410.0, "arena_elo_hard_prompts": 1439.7, "arena_elo_math": 1430.0, "arena_elo_overall": 1400.0, "arena_elo_style_control": 1431.2, "arena_elo_vision": 1217.6, "gpqa_diamond": 87.7, "humaneval": 92.8, "ifeval": 91.0, "live_code_bench": 58.3, "math_500": 96.7, "medqa": 96.1, "mmlu_pro": 84.1, "multipl_e_csharp": 87.8, "multipl_e_julia": 69.8, "multipl_e_kotlin": 82.1, "multipl_e_lua": 65.2, "multipl_e_perl": 62.1, "multipl_e_php": 85.1, "multipl_e_r": 67.5, "multipl_e_ruby": 77.5, "multipl_e_scala": 71.8, "multipl_e_swift": 75.8, "swe_bench_agent": 62.8, "swe_bench_verified": 71.7, "tau_bench": 65.5, "terminal_bench": 58.2}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 200000, "cost_input": 2.0, "display_name": "o3", "fits": {}, "model_id": "openai/o3", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.0, "aime_2025": 87.0, "arena_elo_coding": 1390.0, "arena_elo_math": 1410.0, "arena_elo_overall": 1380.0, "gpqa_diamond": 82.3, "humaneval": 91.2, "ifeval": 89.0, "live_code_bench": 85.9, "math_500": 95.8, "mmlu_pro": 81.0, "multipl_e_csharp": 84.5, "multipl_e_julia": 67.2, "multipl_e_kotlin": 78.2, "multipl_e_lua": 61.5, "multipl_e_perl": 57.2, "multipl_e_php": 82.1, "multipl_e_r": 63.8, "multipl_e_ruby": 71.5, "multipl_e_scala": 66.2, "multipl_e_swift": 71.5, "swe_bench_verified": 68.4}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 2.0, "display_name": "o4-mini-deep-research", "fits": {}, "model_id": "openai/o4-mini-deep-research", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 72.0, "aime_2025": 87.0, "arena_elo_coding": 1390.0, "arena_elo_hard_prompts": 1405.2, "arena_elo_math": 1410.0, "arena_elo_overall": 1380.0, "arena_elo_style_control": 1389.7, "arena_elo_vision": 1201.8, "gpqa_diamond": 82.3, "humaneval": 91.2, "ifeval": 89.0, "live_code_bench": 85.9, "math_500": 95.8, "medqa": 95.2, "mmlu_pro": 81.0, "multipl_e_csharp": 84.5, "multipl_e_julia": 67.2, "multipl_e_kotlin": 78.2, "multipl_e_lua": 61.5, "multipl_e_perl": 57.2, "multipl_e_php": 82.1, "multipl_e_r": 63.8, "multipl_e_ruby": 71.5, "multipl_e_scala": 66.2, "multipl_e_swift": 71.5, "swe_bench_verified": 68.4}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 200000, "cost_input": 1.1, "display_name": "o4-mini", "fits": {}, "model_id": "openai/o4-mini", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 55.1, "miracl": 54.9, "mteb_classification": 70.5, "mteb_clustering": 49.2, "mteb_overall": 64.8, "mteb_pair_classification": 85.5, "mteb_reranking": 59.2, "mteb_retrieval": 62.1, "mteb_sts": 81.5, "mteb_summarization": 30.1}, "capability_tiers": {}, "context_window": 8191, "cost_input": 0.13, "display_name": "text-embedding-3-large", "fits": {}, "model_id": "openai/text-embedding-3-large", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 46.8, "miracl": 42.0, "mteb_classification": 64.8, "mteb_clustering": 42.5, "mteb_overall": 58.5, "mteb_pair_classification": 82.0, "mteb_reranking": 55.0, "mteb_retrieval": 53.2, "mteb_sts": 77.2, "mteb_summarization": 28.5}, "capability_tiers": {}, "context_window": 8191, "cost_input": 0.02, "display_name": "text-embedding-3-small", "fits": {}, "model_id": "openai/text-embedding-3-small", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 43.5, "miracl": 35.5, "mteb_classification": 62.1, "mteb_clustering": 39.8, "mteb_overall": 55.8, "mteb_pair_classification": 80.2, "mteb_reranking": 52.8, "mteb_retrieval": 49.8, "mteb_sts": 74.5, "mteb_summarization": 27.0}, "capability_tiers": {}, "context_window": 8192, "cost_input": 0.1, "display_name": "text-embedding-ada-002", "fits": {}, "model_id": "openai/text-embedding-ada-002", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "OpenAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "MiniCPM 1B sft bf16", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "openbmb/minicpm-1b-sft-bf16", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "MiniCPM 2B sft bf16", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "openbmb/minicpm-2b-sft-bf16", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "MiniCPM Llama3 V 2 5", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-llama3-v-2-5", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM o 2 6", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-o-2-6", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM o 4 5 gguf", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-o-4-5-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM o 4 5", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-o-4-5", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM V 2 6 gguf", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-v-2-6-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM V 2 6 int4", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-v-2-6-int4", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM V 2 6", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-v-2-6", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "MiniCPM V 2", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "openbmb/minicpm-v-2", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {"ai2d": 82.9, "chartqa": 84.4, "docvqa": 92.9, "mathvista": 66.9, "mmmu": 51.2, "ocrbench": 89.4, "realworldqa": 68.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM V 4 5 gguf", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-v-4-5-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "OpenBMB", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 82.9, "chartqa": 84.4, "docvqa": 92.9, "mathvista": 66.9, "mmmu": 51.2, "ocrbench": 89.4, "realworldqa": 68.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM V 4 5", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-v-4-5", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "OpenBMB", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 82.9, "chartqa": 84.4, "docvqa": 92.9, "mathvista": 66.9, "mmmu": 51.2, "ocrbench": 89.4, "realworldqa": 68.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM V 4", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm-v-4", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "OpenBMB", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "MiniCPM3 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "openbmb/minicpm3-4b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "MiniCPM4.1 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "openbmb/minicpm4-1-8b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "OpenBMB", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 128000, "cost_input": 2.0, "display_name": "Perplexity Sonar Deep Research", "fits": {}, "model_id": "perplexity/sonar-deep-research", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Perplexity AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 200000, "cost_input": 3.0, "display_name": "Sonar Pro", "fits": {}, "model_id": "perplexity/sonar-pro", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Perplexity AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 128000, "cost_input": 2.0, "display_name": "Sonar Reasoning Pro", "fits": {}, "model_id": "perplexity/sonar-reasoning-pro", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Perplexity AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 128000, "cost_input": 1.0, "display_name": "Sonar", "fits": {}, "model_id": "perplexity/sonar", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Perplexity AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 1.2, "display_name": "QVQ Max", "fits": {}, "model_id": "qwen/qvq-max", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.05, "display_name": "Qwen Flash", "fits": {}, "model_id": "qwen/qwen-flash", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 1.6, "display_name": "Qwen Max", "fits": {}, "model_id": "qwen/qwen-max", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 16384, "cost_input": 2.46, "display_name": "Qwen-MT Plus", "fits": {}, "model_id": "qwen/qwen-mt-plus", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 16384, "cost_input": 0.16, "display_name": "Qwen-MT Turbo", "fits": {}, "model_id": "qwen/qwen-mt-turbo", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.27, "display_name": "Qwen-Omni Turbo Realtime", "fits": {}, "model_id": "qwen/qwen-omni-turbo-realtime", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.07, "display_name": "Qwen-Omni Turbo", "fits": {}, "model_id": "qwen/qwen-omni-turbo", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 0.5, "display_name": "Qwen Plus Character (Japanese)", "fits": {}, "model_id": "qwen/qwen-plus-character-ja", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.4, "display_name": "Qwen Plus", "fits": {}, "model_id": "qwen/qwen-plus", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.05, "display_name": "Qwen Turbo", "fits": {}, "model_id": "qwen/qwen-turbo", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.8, "display_name": "Qwen-VL Max", "fits": {}, "model_id": "qwen/qwen-vl-max", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 34096, "cost_input": 0.72, "display_name": "Qwen-VL OCR", "fits": {}, "model_id": "qwen/qwen-vl-ocr", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.21, "display_name": "Qwen-VL Plus", "fits": {}, "model_id": "qwen/qwen-vl-plus", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 31.9, "bbh": 31.7, "gpqa_diamond": 24.7, "gsm8k": 35.7, "hellaswag": 48.7, "ifeval": 22.5, "math_500": 2.9, "mmlu_abstract_algebra": 34.0, "mmlu_anatomy": 43.0, "mmlu_astronomy": 41.4, "mmlu_business_ethics": 43.0, "mmlu_clinical_knowledge": 50.2, "mmlu_college_biology": 39.6, "mmlu_college_chemistry": 28.0, "mmlu_college_computer_science": 37.0, "mmlu_college_mathematics": 31.0, "mmlu_college_medicine": 48.0, "mmlu_college_physics": 24.5, "mmlu_computer_security": 50.0, "mmlu_conceptual_physics": 37.0, "mmlu_econometrics": 32.5, "mmlu_electrical_engineering": 47.6, "mmlu_elementary_mathematics": 35.2, "mmlu_formal_logic": 33.3, "mmlu_global_facts": 36.0, "mmlu_high_school_biology": 48.7, "mmlu_high_school_chemistry": 42.4, "mmlu_high_school_computer_science": 47.0, "mmlu_high_school_european_history": 57.6, "mmlu_high_school_geography": 50.5, "mmlu_high_school_government_and_politics": 56.0, "mmlu_high_school_macroeconomics": 45.9, "mmlu_high_school_mathematics": 33.7, "mmlu_high_school_microeconomics": 47.5, "mmlu_high_school_physics": 27.8, "mmlu_high_school_psychology": 56.5, "mmlu_high_school_statistics": 38.9, "mmlu_high_school_us_history": 53.9, "mmlu_high_school_world_history": 59.1, "mmlu_human_aging": 43.0, "mmlu_human_sexuality": 48.1, "mmlu_international_law": 68.6, "mmlu_jurisprudence": 59.3, "mmlu_logical_fallacies": 47.9, "mmlu_machine_learning": 25.9, "mmlu_management": 63.1, "mmlu_marketing": 68.4, "mmlu_medical_genetics": 43.0, "mmlu_miscellaneous": 52.1, "mmlu_moral_disputes": 51.2, "mmlu_moral_scenarios": 25.6, "mmlu_nutrition": 51.6, "mmlu_philosophy": 50.5, "mmlu_prehistory": 43.8, "mmlu_pro": 15.3, "mmlu_professional_accounting": 37.2, "mmlu_professional_law": 36.2, "mmlu_professional_medicine": 44.9, "mmlu_professional_psychology": 41.3, "mmlu_public_relations": 50.0, "mmlu_security_studies": 54.3, "mmlu_sociology": 61.2, "mmlu_us_foreign_policy": 65.0, "mmlu_virology": 37.3, "mmlu_world_religions": 48.5, "musr": 33.5, "truthfulqa": 39.8, "winogrande": 55.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2 0.5B Instruct", "fits": {"amd_rx_7900_xt": 2240.0, "amd_rx_7900_xtx": 2688.0, "amd_rx_9070_xt": 1806.0, "apple_m1_max": 1120.0, "apple_m2_max": 1120.0, "apple_m2_ultra": 2240.0, "apple_m3_max": 1120.0, "apple_m3_ultra": 2293.2, "apple_m4": 336.0, "apple_m4_max": 1528.8, "apple_m4_pro": 764.4, "nvidia_rtx_3060_12gb": 1008.0, "nvidia_rtx_3090": 2620.8, "nvidia_rtx_4060_ti_16gb": 806.4, "nvidia_rtx_4070_ti_super": 1881.6, "nvidia_rtx_4080_super": 2060.8, "nvidia_rtx_4090": 2822.4, "nvidia_rtx_5080": 2688.0, "nvidia_rtx_5090": 5017.6}, "model_id": "qwen/qwen2-0-5b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 44.3, "bbh": 38.5, "gpqa_diamond": 26.2, "gsm8k": 55.8, "hellaswag": 66.7, "ifeval": 33.7, "math_500": 7.2, "mmlu_abstract_algebra": 34.0, "mmlu_anatomy": 48.9, "mmlu_astronomy": 55.3, "mmlu_business_ethics": 67.0, "mmlu_clinical_knowledge": 60.0, "mmlu_college_biology": 56.2, "mmlu_college_chemistry": 38.0, "mmlu_college_computer_science": 51.0, "mmlu_college_mathematics": 32.0, "mmlu_college_medicine": 49.1, "mmlu_college_physics": 40.2, "mmlu_computer_security": 66.0, "mmlu_conceptual_physics": 50.2, "mmlu_econometrics": 38.6, "mmlu_electrical_engineering": 57.9, "mmlu_elementary_mathematics": 42.3, "mmlu_formal_logic": 38.9, "mmlu_global_facts": 36.0, "mmlu_high_school_biology": 64.2, "mmlu_high_school_chemistry": 49.3, "mmlu_high_school_computer_science": 56.0, "mmlu_high_school_european_history": 70.9, "mmlu_high_school_geography": 73.7, "mmlu_high_school_government_and_politics": 78.8, "mmlu_high_school_macroeconomics": 58.5, "mmlu_high_school_mathematics": 35.9, "mmlu_high_school_microeconomics": 62.2, "mmlu_high_school_physics": 35.8, "mmlu_high_school_psychology": 75.2, "mmlu_high_school_statistics": 44.4, "mmlu_high_school_us_history": 64.7, "mmlu_high_school_world_history": 70.5, "mmlu_human_aging": 61.9, "mmlu_human_sexuality": 67.2, "mmlu_international_law": 74.4, "mmlu_jurisprudence": 69.4, "mmlu_logical_fallacies": 68.7, "mmlu_machine_learning": 42.0, "mmlu_management": 83.5, "mmlu_marketing": 80.3, "mmlu_medical_genetics": 62.0, "mmlu_miscellaneous": 72.0, "mmlu_moral_disputes": 62.7, "mmlu_moral_scenarios": 29.7, "mmlu_nutrition": 69.3, "mmlu_philosophy": 64.3, "mmlu_prehistory": 59.0, "mmlu_pro": 25.0, "mmlu_professional_accounting": 43.6, "mmlu_professional_law": 42.8, "mmlu_professional_medicine": 50.4, "mmlu_professional_psychology": 55.4, "mmlu_public_relations": 60.0, "mmlu_security_studies": 68.2, "mmlu_sociology": 75.6, "mmlu_us_foreign_policy": 79.0, "mmlu_virology": 46.4, "mmlu_world_religions": 74.3, "musr": 42.9, "truthfulqa": 45.9, "winogrande": 64.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2 1.5B Instruct", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "qwen/qwen2-1-5b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 33.2, "gpqa_diamond": 25.9, "ifeval": 31.5, "math_500": 10.3, "mmlu_pro": 17.2, "musr": 33.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 0.5B Instruct", "fits": {"amd_rx_7900_xt": 2240.0, "amd_rx_7900_xtx": 2688.0, "amd_rx_9070_xt": 1806.0, "apple_m1_max": 1120.0, "apple_m2_max": 1120.0, "apple_m2_ultra": 2240.0, "apple_m3_max": 1120.0, "apple_m3_ultra": 2293.2, "apple_m4": 336.0, "apple_m4_max": 1528.8, "apple_m4_pro": 764.4, "nvidia_rtx_3060_12gb": 1008.0, "nvidia_rtx_3090": 2620.8, "nvidia_rtx_4060_ti_16gb": 806.4, "nvidia_rtx_4070_ti_super": 1881.6, "nvidia_rtx_4080_super": 2060.8, "nvidia_rtx_4090": 2822.4, "nvidia_rtx_5080": 2688.0, "nvidia_rtx_5090": 5017.6}, "model_id": "qwen/qwen2-5-0-5b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 32.7, "gpqa_diamond": 24.7, "ifeval": 16.3, "math_500": 3.9, "mmlu_pro": 19.1, "musr": 34.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 0.5B", "fits": {"amd_rx_7900_xt": 2240.0, "amd_rx_7900_xtx": 2688.0, "amd_rx_9070_xt": 1806.0, "apple_m1_max": 1120.0, "apple_m2_max": 1120.0, "apple_m2_ultra": 2240.0, "apple_m3_max": 1120.0, "apple_m3_ultra": 2293.2, "apple_m4": 336.0, "apple_m4_max": 1528.8, "apple_m4_pro": 764.4, "nvidia_rtx_3060_12gb": 1008.0, "nvidia_rtx_3090": 2620.8, "nvidia_rtx_4060_ti_16gb": 806.4, "nvidia_rtx_4070_ti_super": 1881.6, "nvidia_rtx_4080_super": 2060.8, "nvidia_rtx_4090": 2822.4, "nvidia_rtx_5080": 2688.0, "nvidia_rtx_5090": 5017.6}, "model_id": "qwen/qwen2-5-0-5b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 40.8, "gpqa_diamond": 28.5, "ifeval": 26.7, "math_500": 9.1, "mmlu_pro": 28.5, "musr": 35.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 1.5B Instruct AWQ", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "qwen/qwen2-5-1-5b-instruct-awq", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 42.9, "gpqa_diamond": 25.6, "ifeval": 44.8, "math_500": 22.1, "mmlu_pro": 28.0, "musr": 36.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 1.5B Instruct", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "qwen/qwen2-5-1-5b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 40.8, "gpqa_diamond": 28.5, "ifeval": 26.7, "math_500": 9.1, "mmlu_pro": 28.5, "musr": 35.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 1.5B", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "qwen/qwen2-5-1-5b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 51.2, "gpqa_diamond": 42.1, "humaneval": 72.5, "ifeval": 74.0, "math_500": 68.5, "mmlu_pro": 58.2, "musr": 18.6}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 14B Instruct AWQ", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "qwen/qwen2-5-14b-instruct-awq", "model_subtypes": ["llm-code"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 51.2, "gpqa_diamond": 42.1, "humaneval": 72.5, "ifeval": 74.0, "math_500": 68.5, "mmlu_pro": 58.2, "musr": 18.6}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.35, "display_name": "Qwen2.5 14B Instruct", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "qwen/qwen2-5-14b-instruct", "model_subtypes": ["llm-code"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1240.0, "arena_elo_overall": 1250.0, "bbh": 58.4, "gpqa_diamond": 48.5, "humaneval": 78.2, "ifeval": 78.0, "math_500": 76.8, "mmlu_pro": 65.2, "musr": 22.1}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 32B Instruct AWQ", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "qwen/qwen2-5-32b-instruct-awq", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1240.0, "arena_elo_overall": 1250.0, "bbh": 58.4, "gpqa_diamond": 48.5, "humaneval": 78.2, "ifeval": 78.0, "math_500": 76.8, "mmlu_pro": 65.2, "musr": 22.1}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.7, "display_name": "Qwen2.5 32B Instruct", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "qwen/qwen2-5-32b-instruct", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 46.9, "gpqa_diamond": 27.3, "ifeval": 64.7, "math_500": 36.8, "mmlu_pro": 32.5, "musr": 39.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 3B Instruct", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "qwen/qwen2-5-3b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {"alpaca_eval": 38.1, "arena_elo_coding": 1290.0, "arena_elo_hard_prompts": 1317.3, "arena_elo_math": 1270.0, "arena_elo_overall": 1280.0, "arena_elo_style_control": 1302.3, "arena_elo_vision": 1122.0, "bbh": 66.3, "bbq": 75.2, "gpqa_diamond": 54.8, "helm_safety": 80.5, "humaneval": 84.5, "ifeval": 82.0, "live_code_bench": 55.5, "math_500": 83.2, "mmlu_pro": 71.5, "multipl_e_csharp": 74.5, "multipl_e_julia": 54.8, "multipl_e_kotlin": 65.5, "multipl_e_lua": 49.5, "multipl_e_perl": 44.2, "multipl_e_php": 72.2, "multipl_e_r": 50.5, "multipl_e_ruby": 58.5, "multipl_e_scala": 53.2, "multipl_e_swift": 58.8, "musr": 28.3, "toxigen": 84.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 1.4, "display_name": "Qwen2.5 72B Instruct", "fits": {"apple_m1_max": 7.8, "apple_m2_max": 7.8, "apple_m2_ultra": 15.6, "apple_m3_max": 7.8, "apple_m3_ultra": 15.9, "apple_m4_max": 10.6, "apple_m4_pro": 5.3}, "model_id": "qwen/qwen2-5-72b-instruct", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 43.8, "gpqa_diamond": 35.2, "humaneval": 65.0, "ifeval": 68.0, "math_500": 58.5, "mmlu_pro": 50.1, "musr": 14.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 7B Instruct AWQ", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-5-7b-instruct-awq", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 43.8, "gpqa_diamond": 35.2, "humaneval": 65.0, "ifeval": 68.0, "math_500": 58.5, "mmlu_pro": 50.1, "musr": 14.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.175, "display_name": "Qwen2.5 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-5-7b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 43.8, "gpqa_diamond": 35.2, "gsm8k": 91.6, "humaneval": 65.0, "ifeval": 68.0, "math_500": 58.5, "mmlu_pro": 50.1, "musr": 14.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-5-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 Coder 1.5B Instruct", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "qwen/qwen2-5-coder-1-5b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"bbh": 61.4, "gpqa_diamond": 30.5, "ifeval": 69.1, "math_500": 32.5, "mmlu_pro": 39.4, "multipl_e": 74.1, "multipl_e_cpp": 73.2, "multipl_e_csharp": 68.2, "multipl_e_go": 70.1, "multipl_e_java": 78.5, "multipl_e_javascript": 76.5, "multipl_e_kotlin": 56.8, "multipl_e_lua": 45.5, "multipl_e_php": 66.5, "multipl_e_python": 82.1, "multipl_e_ruby": 50.2, "multipl_e_rust": 62.5, "multipl_e_scala": 47.5, "multipl_e_swift": 52.1, "multipl_e_typescript": 75.8, "musr": 39.1}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 Coder 14B Instruct", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "qwen/qwen2-5-coder-14b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 16.4, "bbh": 64.0, "gpqa_diamond": 42.5, "humaneval": 87.2, "ifeval": 76.0, "live_code_bench": 29.5, "math_500": 70.5, "mmlu_pro": 60.2, "multipl_e": 80.3, "multipl_e_cpp": 79.8, "multipl_e_csharp": 76.5, "multipl_e_go": 76.5, "multipl_e_java": 83.5, "multipl_e_javascript": 82.1, "multipl_e_kotlin": 64.8, "multipl_e_lua": 52.8, "multipl_e_php": 74.2, "multipl_e_python": 88.5, "multipl_e_ruby": 58.5, "multipl_e_rust": 70.2, "multipl_e_scala": 55.2, "multipl_e_swift": 60.5, "multipl_e_typescript": 81.2, "musr": 45.3}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 Coder 32B Instruct AWQ", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "qwen/qwen2-5-coder-32b-instruct-awq", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 16.4, "bbh": 66.3, "gpqa_diamond": 42.5, "humaneval": 87.2, "ifeval": 76.0, "live_code_bench": 29.5, "math_500": 70.5, "mmlu_pro": 60.2, "multipl_e": 80.3, "multipl_e_cpp": 79.8, "multipl_e_csharp": 76.5, "multipl_e_go": 76.5, "multipl_e_java": 83.5, "multipl_e_javascript": 82.1, "multipl_e_kotlin": 64.8, "multipl_e_lua": 52.8, "multipl_e_php": 74.2, "multipl_e_python": 88.5, "multipl_e_ruby": 58.5, "multipl_e_rust": 70.2, "multipl_e_scala": 55.2, "multipl_e_swift": 60.5, "multipl_e_typescript": 81.2, "musr": 43.9}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 Coder 32B Instruct", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "qwen/qwen2-5-coder-32b-instruct", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 48.6, "gpqa_diamond": 25.9, "ifeval": 34.5, "math_500": 19.2, "mmlu_pro": 36.8, "multipl_e": 67.6, "multipl_e_cpp": 66.5, "multipl_e_csharp": 60.8, "multipl_e_go": 63.8, "multipl_e_java": 72.1, "multipl_e_javascript": 70.2, "multipl_e_kotlin": 48.5, "multipl_e_lua": 38.5, "multipl_e_php": 59.0, "multipl_e_python": 75.8, "multipl_e_ruby": 42.5, "multipl_e_rust": 55.2, "multipl_e_scala": 40.2, "multipl_e_swift": 44.2, "multipl_e_typescript": 69.5, "musr": 34.5}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 Coder 7B Instruct GPTQ Int4", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-5-coder-7b-instruct-gptq-int4", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 50.0, "gpqa_diamond": 29.4, "ifeval": 61.5, "math_500": 3.1, "mmlu_pro": 33.5, "multipl_e": 67.6, "multipl_e_cpp": 66.5, "multipl_e_csharp": 60.8, "multipl_e_go": 63.8, "multipl_e_java": 72.1, "multipl_e_javascript": 70.2, "multipl_e_kotlin": 48.5, "multipl_e_lua": 38.5, "multipl_e_php": 59.0, "multipl_e_python": 75.8, "multipl_e_ruby": 42.5, "multipl_e_rust": 55.2, "multipl_e_scala": 40.2, "multipl_e_swift": 44.2, "multipl_e_typescript": 69.5, "musr": 41.0}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 Coder 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-5-coder-7b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 37.5, "gpqa_diamond": 26.5, "ifeval": 18.6, "math_500": 26.3, "mmlu_pro": 18.0, "musr": 36.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Qwen2.5 Math 1.5B", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "qwen/qwen2-5-math-1-5b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Qwen2.5-Omni 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-5-omni-7b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"chartqa": 72.5, "docvqa": 85.2, "mathvista": 45.8, "mmmu": 42.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2.5 VL 3B Instruct", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "qwen/qwen2-5-vl-3b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 88.7, "bbq": 79.5, "chartqa": 82.5, "docvqa": 90.1, "helm_safety": 84.2, "mathvista": 58.2, "mmmu": 64.5, "ocrbench": 88.5, "toxigen": 87.8}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 2.8, "display_name": "Qwen2.5-VL 72B Instruct", "fits": {"apple_m1_max": 7.8, "apple_m2_max": 7.8, "apple_m2_ultra": 15.6, "apple_m3_max": 7.8, "apple_m3_ultra": 15.9, "apple_m4_max": 10.6, "apple_m4_pro": 5.3}, "model_id": "qwen/qwen2-5-vl-72b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"ai2d": 79.5, "chartqa": 65.8, "docvqa": 78.5, "mathvista": 42.5, "mmmu": 48.2, "ocrbench": 86.4}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.35, "display_name": "Qwen2.5-VL 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-5-vl-7b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2 VL 2B Instruct", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "qwen/qwen2-vl-2b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"ai2d": 83.0, "bbh": 54.6, "chartqa": 83.0, "docvqa": 94.5, "gpqa_diamond": 32.0, "ifeval": 46.0, "math_500": 19.9, "mathvista": 58.2, "mmlu_pro": 40.9, "mmmu": 54.1, "musr": 43.8}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2 VL 7B Instruct AWQ", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-vl-7b-instruct-awq", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {"ai2d": 83.0, "bbh": 54.6, "chartqa": 83.0, "docvqa": 94.5, "gpqa_diamond": 32.0, "ifeval": 46.0, "math_500": 19.9, "mathvista": 58.2, "mmlu_pro": 40.9, "mmmu": 54.1, "musr": 43.8}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen2 VL 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "qwen/qwen2-vl-7b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.03, "display_name": "Qwen3 0.6B FP8", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "qwen/qwen3-0-6b-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.03, "display_name": "Qwen3 0.6B", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "qwen/qwen3-0-6b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.04, "display_name": "Qwen3 1.7B", "fits": {"amd_rx_7900_xt": 658.8, "amd_rx_7900_xtx": 790.6, "amd_rx_9070_xt": 531.2, "apple_m1_max": 329.4, "apple_m2_max": 329.4, "apple_m2_ultra": 658.8, "apple_m3_max": 329.4, "apple_m3_ultra": 674.5, "apple_m4": 98.8, "apple_m4_max": 449.6, "apple_m4_pro": 224.8, "nvidia_rtx_3060_12gb": 296.5, "nvidia_rtx_3090": 770.8, "nvidia_rtx_4060_ti_16gb": 237.2, "nvidia_rtx_4070_ti_super": 553.4, "nvidia_rtx_4080_super": 606.1, "nvidia_rtx_4090": 830.1, "nvidia_rtx_5080": 790.6, "nvidia_rtx_5090": 1475.8}, "model_id": "qwen/qwen3-1-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1280.0, "arena_elo_math": 1290.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 70.0, "artificial_analysis_speed_index": 80.0, "bbh": 55.1, "gpqa_diamond": 50.1, "humaneval": 76.5, "ifeval": 77.0, "math_500": 80.5, "mmlu_pro": 64.2, "musr": 24.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.2, "display_name": "Qwen3 14B AWQ", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "qwen/qwen3-14b-awq", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1280.0, "arena_elo_math": 1290.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 70.0, "artificial_analysis_speed_index": 80.0, "bbh": 55.1, "gpqa_diamond": 50.1, "humaneval": 76.5, "ifeval": 77.0, "math_500": 80.5, "mmlu_pro": 64.2, "musr": 24.8}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.35, "display_name": "Qwen3 14B", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "qwen/qwen3-14b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 68.2, "alpaca_eval": 44.5, "arena_elo_coding": 1350.0, "arena_elo_hard_prompts": 1440.4, "arena_elo_math": 1360.0, "arena_elo_overall": 1340.0, "arena_elo_style_control": 1415.8, "arena_elo_vision": 1214.6, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 55.0, "gpqa_diamond": 71.1, "gsm8k": 94.4, "humaneval": 89.5, "ifeval": 88.5, "live_code_bench": 62.2, "math_500": 93.5, "medqa": 84.8, "mgsm": 83.5, "mmlu_astronomy": 75.8, "mmlu_biology": 83.8, "mmlu_business_ethics": 76.5, "mmlu_chemistry": 77.2, "mmlu_clinical_knowledge": 82.2, "mmlu_computer_science": 84.5, "mmlu_jurisprudence": 73.8, "mmlu_physics": 81.5, "mmlu_pro": 80.2, "mmlu_professional_accounting": 67.5, "mmlu_professional_law": 71.8, "mt_bench": 8.9, "multipl_e_csharp": 83.5, "multipl_e_julia": 65.8, "multipl_e_kotlin": 76.8, "multipl_e_lua": 58.8, "multipl_e_perl": 55.2, "multipl_e_php": 81.2, "multipl_e_r": 62.5, "multipl_e_ruby": 70.5, "multipl_e_scala": 65.2, "multipl_e_swift": 70.2, "swe_bench_agent": 45.5, "swe_bench_verified": 55.8, "tau_bench": 50.2, "wildbench": 73.8}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.7, "display_name": "Qwen3 235B-A22B", "fits": {"apple_m2_ultra": 4.8, "apple_m3_ultra": 4.9}, "model_id": "qwen/qwen3-235b-a22b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1280.0, "artificial_analysis_quality_index": 74.0, "artificial_analysis_speed_index": 88.0, "bbh": 50.5, "gpqa_diamond": 55.2, "humaneval": 80.5, "ifeval": 80.0, "live_code_bench": 62.6, "math_500": 84.2, "mmlu_pro": 68.5, "musr": 21.3}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.15, "display_name": "Qwen3 30B A3B Instruct 2507", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "qwen/qwen3-30b-a3b-instruct-2507", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_hard_prompts": 1345.6, "arena_elo_math": 1300.0, "arena_elo_overall": 1280.0, "arena_elo_style_control": 1327.3, "artificial_analysis_quality_index": 74.0, "artificial_analysis_speed_index": 88.0, "bbh": 50.5, "gpqa_diamond": 55.2, "humaneval": 80.5, "ifeval": 80.0, "live_code_bench": 62.6, "math_500": 84.2, "medqa": 85.3, "mmlu_pro": 68.5, "musr": 21.3}, "capability_tiers": {"coding": "tier-1", "coding:agentic_coding": "tier-1", "coding:code_review": "tier-1", "coding:debugging": "tier-1", "coding:documentation": "tier-1", "coding:refactoring": "tier-1", "coding:test_generation": "tier-1", "language:multilingual": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:logical": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "reasoning:planning": "tier-1", "reasoning:self_correction": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1", "tool_use:mcp_compatible": "tier-1", "tool_use:parallel_tool_calls": "tier-1"}, "context_window": 32768, "cost_input": 0.0, "display_name": "Qwen3 30B-A3B", "fits": {}, "model_id": "qwen/qwen3-30b-a3b", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": null, "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 40.0, "alpaca_eval": 38.2, "arena_elo_coding": 1320.0, "arena_elo_math": 1330.0, "arena_elo_overall": 1310.0, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 70.0, "bbh": 62.8, "gpqa_diamond": 62.5, "humaneval": 85.2, "ifeval": 84.0, "live_code_bench": 54.6, "math_500": 88.5, "mmlu_pro": 74.8, "mt_bench": 8.5, "multipl_e_csharp": 78.2, "multipl_e_julia": 59.2, "multipl_e_kotlin": 70.5, "multipl_e_lua": 53.5, "multipl_e_perl": 48.5, "multipl_e_php": 75.8, "multipl_e_r": 55.5, "multipl_e_ruby": 63.5, "multipl_e_scala": 58.2, "multipl_e_swift": 63.8, "musr": 30.1, "wildbench": 65.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.3, "display_name": "Qwen3 32B AWQ", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "qwen/qwen3-32b-awq", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 40.0, "alpaca_eval": 38.2, "arena_elo_coding": 1320.0, "arena_elo_hard_prompts": 1367.6, "arena_elo_math": 1330.0, "arena_elo_overall": 1310.0, "arena_elo_style_control": 1347.0, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 70.0, "bbh": 62.8, "gpqa_diamond": 62.5, "humaneval": 85.2, "ifeval": 84.0, "live_code_bench": 54.6, "math_500": 88.5, "mmlu_pro": 74.8, "mt_bench": 8.5, "multipl_e_csharp": 78.2, "multipl_e_julia": 59.2, "multipl_e_kotlin": 70.5, "multipl_e_lua": 53.5, "multipl_e_perl": 48.5, "multipl_e_php": 75.8, "multipl_e_r": 55.5, "multipl_e_ruby": 63.5, "multipl_e_scala": 58.2, "multipl_e_swift": 63.8, "musr": 30.1, "wildbench": 65.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.7, "display_name": "Qwen3 32B", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "qwen/qwen3-32b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1220.0, "arena_elo_overall": 1200.0, "gpqa_diamond": 34.2, "humaneval": 60.5, "ifeval": 66.0, "math_500": 65.3, "mmlu_pro": 48.5}, "capability_tiers": {"reasoning": "tier-2"}, "context_window": 32768, "cost_input": 0.05, "display_name": "Qwen3 4B Base", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-4b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1220.0, "arena_elo_overall": 1200.0, "gpqa_diamond": 34.2, "humaneval": 60.5, "ifeval": 66.0, "math_500": 65.3, "mmlu_pro": 48.5}, "capability_tiers": {"reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.05, "display_name": "Qwen3 4B Instruct 2507 FP8", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-4b-instruct-2507-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1220.0, "arena_elo_overall": 1200.0, "gpqa_diamond": 34.2, "humaneval": 60.5, "ifeval": 66.0, "math_500": 65.3, "mmlu_pro": 48.5}, "capability_tiers": {"reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.05, "display_name": "Qwen3 4B Instruct 2507", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-4b-instruct-2507", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 32768, "cost_input": 0.05, "display_name": "Qwen3 4B Thinking 2507", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-4b-thinking-2507", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1210.0, "arena_elo_math": 1220.0, "arena_elo_overall": 1200.0, "gpqa_diamond": 34.2, "humaneval": 60.5, "ifeval": 66.0, "math_500": 65.3, "mmlu_pro": 48.5}, "capability_tiers": {"reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.05, "display_name": "Qwen3 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-4b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen3.5 0.8B", "fits": {"amd_rx_7900_xt": 1400.0, "amd_rx_7900_xtx": 1680.0, "amd_rx_9070_xt": 1128.7, "apple_m1_max": 700.0, "apple_m2_max": 700.0, "apple_m2_ultra": 1400.0, "apple_m3_max": 700.0, "apple_m3_ultra": 1433.2, "apple_m4": 210.0, "apple_m4_max": 955.5, "apple_m4_pro": 477.7, "nvidia_rtx_3060_12gb": 630.0, "nvidia_rtx_3090": 1638.0, "nvidia_rtx_4060_ti_16gb": 504.0, "nvidia_rtx_4070_ti_super": 1176.0, "nvidia_rtx_4080_super": 1288.0, "nvidia_rtx_4090": 1764.0, "nvidia_rtx_5080": 1680.0, "nvidia_rtx_5090": 3136.0}, "model_id": "qwen/qwen3-5-0-8b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.55, "display_name": "Qwen3.5 122B A10B FP8", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.2, "apple_m3_max": 4.6, "apple_m3_ultra": 9.4, "apple_m4_max": 6.3}, "model_id": "qwen/qwen3-5-122b-a10b-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.55, "display_name": "Qwen3.5 122B A10B", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.2, "apple_m3_max": 4.6, "apple_m3_ultra": 9.4, "apple_m4_max": 6.3}, "model_id": "qwen/qwen3-5-122b-a10b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.25, "display_name": "Qwen3.5 27B FP8", "fits": {"amd_rx_7900_xt": 41.5, "amd_rx_7900_xtx": 49.8, "apple_m1_max": 20.7, "apple_m2_max": 20.7, "apple_m2_ultra": 41.5, "apple_m3_max": 20.7, "apple_m3_ultra": 42.5, "apple_m4": 6.2, "apple_m4_max": 28.3, "apple_m4_pro": 14.2, "nvidia_rtx_3090": 48.5, "nvidia_rtx_4090": 52.3, "nvidia_rtx_5090": 92.9}, "model_id": "qwen/qwen3-5-27b-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.25, "display_name": "Qwen3.5 27B", "fits": {"amd_rx_7900_xt": 41.5, "amd_rx_7900_xtx": 49.8, "apple_m1_max": 20.7, "apple_m2_max": 20.7, "apple_m2_ultra": 41.5, "apple_m3_max": 20.7, "apple_m3_ultra": 42.5, "apple_m4": 6.2, "apple_m4_max": 28.3, "apple_m4_pro": 14.2, "nvidia_rtx_3090": 48.5, "nvidia_rtx_4090": 52.3, "nvidia_rtx_5090": 92.9}, "model_id": "qwen/qwen3-5-27b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.04, "display_name": "Qwen3.5 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "qwen/qwen3-5-2b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.15, "display_name": "Qwen3.5 35B A3B FP8", "fits": {"amd_rx_7900_xtx": 38.4, "apple_m1_max": 16.0, "apple_m2_max": 16.0, "apple_m2_ultra": 32.0, "apple_m3_max": 16.0, "apple_m3_ultra": 32.8, "apple_m4": 4.8, "apple_m4_max": 21.8, "apple_m4_pro": 10.9, "nvidia_rtx_3090": 37.4, "nvidia_rtx_4090": 40.3, "nvidia_rtx_5090": 71.7}, "model_id": "qwen/qwen3-5-35b-a3b-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.15, "display_name": "Qwen3.5 35B A3B GPTQ Int4", "fits": {"amd_rx_7900_xtx": 38.4, "apple_m1_max": 16.0, "apple_m2_max": 16.0, "apple_m2_ultra": 32.0, "apple_m3_max": 16.0, "apple_m3_ultra": 32.8, "apple_m4": 4.8, "apple_m4_max": 21.8, "apple_m4_pro": 10.9, "nvidia_rtx_3090": 37.4, "nvidia_rtx_4090": 40.3, "nvidia_rtx_5090": 71.7}, "model_id": "qwen/qwen3-5-35b-a3b-gptq-int4", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.15, "display_name": "Qwen3.5 35B A3B", "fits": {"amd_rx_7900_xtx": 38.4, "apple_m1_max": 16.0, "apple_m2_max": 16.0, "apple_m2_ultra": 32.0, "apple_m3_max": 16.0, "apple_m3_ultra": 32.8, "apple_m4": 4.8, "apple_m4_max": 21.8, "apple_m4_pro": 10.9, "nvidia_rtx_3090": 37.4, "nvidia_rtx_4090": 40.3, "nvidia_rtx_5090": 71.7}, "model_id": "qwen/qwen3-5-35b-a3b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 1.2, "display_name": "Qwen3.5 397B A17B FP8", "fits": {"apple_m3_ultra": 2.9}, "model_id": "qwen/qwen3-5-397b-a17b-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.6, "display_name": "Qwen3.5 397B-A17B", "fits": {"apple_m3_ultra": 2.9}, "model_id": "qwen/qwen3-5-397b-a17b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.05, "display_name": "Qwen3.5 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-5-4b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.1, "display_name": "Qwen3.5 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "qwen/qwen3-5-9b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.4, "display_name": "Qwen3.5 Plus", "fits": {}, "model_id": "qwen/qwen3-5-plus", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.276, "display_name": "Qwen3.6 Plus", "fits": {}, "model_id": "qwen/qwen3-6-plus", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1250.0, "arena_elo_math": 1260.0, "arena_elo_overall": 1240.0, "artificial_analysis_quality_index": 66.0, "artificial_analysis_speed_index": 88.0, "bbh": 48.3, "gpqa_diamond": 42.5, "humaneval": 70.2, "ifeval": 73.0, "math_500": 74.8, "mmlu_pro": 57.8, "musr": 19.4}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Qwen3 8B AWQ", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "qwen/qwen3-8b-awq", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1250.0, "arena_elo_math": 1260.0, "arena_elo_overall": 1240.0, "artificial_analysis_quality_index": 66.0, "artificial_analysis_speed_index": 88.0, "bbh": 48.3, "gpqa_diamond": 42.5, "humaneval": 70.2, "ifeval": 73.0, "math_500": 74.8, "mmlu_pro": 57.8, "musr": 19.4}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "Qwen3 8B Base", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "qwen/qwen3-8b-base", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-base", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1250.0, "arena_elo_math": 1260.0, "arena_elo_overall": 1240.0, "artificial_analysis_quality_index": 66.0, "artificial_analysis_speed_index": 88.0, "bbh": 48.3, "gpqa_diamond": 42.5, "humaneval": 70.2, "ifeval": 73.0, "math_500": 74.8, "mmlu_pro": 57.8, "musr": 19.4}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.18, "display_name": "Qwen3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "qwen/qwen3-8b", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 ASR 1.7B", "fits": {"amd_rx_7900_xt": 658.8, "amd_rx_7900_xtx": 790.6, "amd_rx_9070_xt": 531.2, "apple_m1_max": 329.4, "apple_m2_max": 329.4, "apple_m2_ultra": 658.8, "apple_m3_max": 329.4, "apple_m3_ultra": 674.5, "apple_m4": 98.8, "apple_m4_max": 449.6, "apple_m4_pro": 224.8, "nvidia_rtx_3060_12gb": 296.5, "nvidia_rtx_3090": 770.8, "nvidia_rtx_4060_ti_16gb": 237.2, "nvidia_rtx_4070_ti_super": 553.4, "nvidia_rtx_4080_super": 606.1, "nvidia_rtx_4090": 830.1, "nvidia_rtx_5080": 790.6, "nvidia_rtx_5090": 1475.8}, "model_id": "qwen/qwen3-asr-1-7b", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 53248, "cost_input": 0.035, "display_name": "Qwen3-ASR Flash", "fits": {}, "model_id": "qwen/qwen3-asr-flash", "model_subtypes": [], "model_type": "audio-asr", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.45, "display_name": "Qwen3-Coder 30B-A3B Instruct", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "qwen/qwen3-coder-30b-a3b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"aider_polyglot": 72.5, "arena_elo_coding": 1380.0, "arena_elo_hard_prompts": 1413.6, "arena_elo_math": 1340.0, "arena_elo_overall": 1350.0, "arena_elo_style_control": 1387.4, "gpqa_diamond": 70.5, "humaneval": 92.5, "ifeval": 88.0, "live_code_bench": 58.5, "math_500": 91.2, "mmlu_pro": 78.5, "multipl_e_csharp": 86.5, "multipl_e_julia": 69.2, "multipl_e_kotlin": 80.2, "multipl_e_lua": 62.5, "multipl_e_perl": 58.8, "multipl_e_php": 84.2, "multipl_e_r": 66.5, "multipl_e_ruby": 74.8, "multipl_e_scala": 69.5, "multipl_e_swift": 74.2, "swe_bench_agent": 52.1, "swe_bench_verified": 68.2, "tau_bench": 55.8, "terminal_bench_2": 37.5}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "reasoning": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 262144, "cost_input": 1.5, "display_name": "Qwen3-Coder 480B-A35B Instruct", "fits": {"apple_m3_ultra": 2.4}, "model_id": "qwen/qwen3-coder-480b-a35b-instruct", "model_subtypes": ["llm-reasoning"], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1000000, "cost_input": 0.3, "display_name": "Qwen3 Coder Flash", "fits": {}, "model_id": "qwen/qwen3-coder-flash", "model_subtypes": [], "model_type": "llm-code", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 262144, "cost_input": null, "display_name": "Qwen3 Coder Next FP8", "fits": {"apple_m3_ultra": 2.4}, "model_id": "qwen/qwen3-coder-next-fp8", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 262144, "cost_input": null, "display_name": "Qwen3 Coder Next", "fits": {"apple_m3_ultra": 2.4}, "model_id": "qwen/qwen3-coder-next", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": 1.0, "display_name": "Qwen3 Coder Plus", "fits": {}, "model_id": "qwen/qwen3-coder-plus", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"beir": 49.5, "miracl": 51.0, "mteb_classification": 66.0, "mteb_clustering": 44.0, "mteb_overall": 60.0, "mteb_pair_classification": 82.0, "mteb_reranking": 60.0, "mteb_retrieval": 56.5, "mteb_sts": 77.5, "mteb_summarization": 28.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "Qwen3 Embedding 0.6B", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "qwen/qwen3-embedding-0-6b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 55.0, "miracl": 57.0, "mteb_classification": 70.5, "mteb_clustering": 49.5, "mteb_overall": 65.5, "mteb_pair_classification": 85.0, "mteb_reranking": 65.1, "mteb_retrieval": 62.5, "mteb_sts": 80.9, "mteb_summarization": 30.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "Qwen3 Embedding 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-embedding-4b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 57.5, "miracl": 60.0, "mteb_classification": 72.5, "mteb_clustering": 52.0, "mteb_overall": 68.0, "mteb_pair_classification": 85.5, "mteb_reranking": 65.6, "mteb_retrieval": 65.5, "mteb_sts": 81.1, "mteb_summarization": 30.5}, "capability_tiers": {}, "context_window": null, "cost_input": 0.008, "display_name": "Qwen3 Embedding 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "qwen/qwen3-embedding-8b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 53248, "cost_input": 10.0, "display_name": "Qwen3-LiveTranslate Flash Realtime", "fits": {}, "model_id": "qwen/qwen3-livetranslate-flash-realtime", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"medqa": 85.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 1.2, "display_name": "Qwen3 Max", "fits": {}, "model_id": "qwen/qwen3-max", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.5, "display_name": "Qwen3-Next 80B-A3B Instruct", "fits": {"apple_m1_max": 7.0, "apple_m2_max": 7.0, "apple_m2_ultra": 14.0, "apple_m3_max": 7.0, "apple_m3_ultra": 14.3, "apple_m4_max": 9.6, "apple_m4_pro": 4.8}, "model_id": "qwen/qwen3-next-80b-a3b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.5, "display_name": "Qwen3-Next 80B-A3B (Thinking)", "fits": {"apple_m1_max": 7.0, "apple_m2_max": 7.0, "apple_m2_ultra": 14.0, "apple_m3_max": 7.0, "apple_m3_ultra": 14.3, "apple_m4_max": 9.6, "apple_m4_pro": 4.8}, "model_id": "qwen/qwen3-next-80b-a3b-thinking", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 65536, "cost_input": 0.52, "display_name": "Qwen3-Omni Flash Realtime", "fits": {}, "model_id": "qwen/qwen3-omni-flash-realtime", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 65536, "cost_input": 0.43, "display_name": "Qwen3-Omni Flash", "fits": {}, "model_id": "qwen/qwen3-omni-flash", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 Reranker 0.6B", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "qwen/qwen3-reranker-0-6b", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 Reranker 4B", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-reranker-4b", "model_subtypes": [], "model_type": "reranker", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 TTS 12Hz 0.6B Base", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "qwen/qwen3-tts-12hz-0-6b-base", "model_subtypes": [], "model_type": "audio-tts", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 TTS 12Hz 1.7B Base", "fits": {"amd_rx_7900_xt": 658.8, "amd_rx_7900_xtx": 790.6, "amd_rx_9070_xt": 531.2, "apple_m1_max": 329.4, "apple_m2_max": 329.4, "apple_m2_ultra": 658.8, "apple_m3_max": 329.4, "apple_m3_ultra": 674.5, "apple_m4": 98.8, "apple_m4_max": 449.6, "apple_m4_pro": 224.8, "nvidia_rtx_3060_12gb": 296.5, "nvidia_rtx_3090": 770.8, "nvidia_rtx_4060_ti_16gb": 237.2, "nvidia_rtx_4070_ti_super": 553.4, "nvidia_rtx_4080_super": 606.1, "nvidia_rtx_4090": 830.1, "nvidia_rtx_5080": 790.6, "nvidia_rtx_5090": 1475.8}, "model_id": "qwen/qwen3-tts-12hz-1-7b-base", "model_subtypes": [], "model_type": "audio-tts", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 TTS 12Hz 1.7B CustomVoice", "fits": {"amd_rx_7900_xt": 658.8, "amd_rx_7900_xtx": 790.6, "amd_rx_9070_xt": 531.2, "apple_m1_max": 329.4, "apple_m2_max": 329.4, "apple_m2_ultra": 658.8, "apple_m3_max": 329.4, "apple_m3_ultra": 674.5, "apple_m4": 98.8, "apple_m4_max": 449.6, "apple_m4_pro": 224.8, "nvidia_rtx_3060_12gb": 296.5, "nvidia_rtx_3090": 770.8, "nvidia_rtx_4060_ti_16gb": 237.2, "nvidia_rtx_4070_ti_super": 553.4, "nvidia_rtx_4080_super": 606.1, "nvidia_rtx_4090": 830.1, "nvidia_rtx_5080": 790.6, "nvidia_rtx_5090": 1475.8}, "model_id": "qwen/qwen3-tts-12hz-1-7b-customvoice", "model_subtypes": [], "model_type": "audio-tts", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 TTS 12Hz 1.7B VoiceDesign", "fits": {"amd_rx_7900_xt": 658.8, "amd_rx_7900_xtx": 790.6, "amd_rx_9070_xt": 531.2, "apple_m1_max": 329.4, "apple_m2_max": 329.4, "apple_m2_ultra": 658.8, "apple_m3_max": 329.4, "apple_m3_ultra": 674.5, "apple_m4": 98.8, "apple_m4_max": 449.6, "apple_m4_pro": 224.8, "nvidia_rtx_3060_12gb": 296.5, "nvidia_rtx_3090": 770.8, "nvidia_rtx_4060_ti_16gb": 237.2, "nvidia_rtx_4070_ti_super": 553.4, "nvidia_rtx_4080_super": 606.1, "nvidia_rtx_4090": 830.1, "nvidia_rtx_5080": 790.6, "nvidia_rtx_5090": 1475.8}, "model_id": "qwen/qwen3-tts-12hz-1-7b-voicedesign", "model_subtypes": [], "model_type": "audio-tts", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.7, "display_name": "Qwen3-VL 235B-A22B", "fits": {"apple_m2_ultra": 4.8, "apple_m3_ultra": 4.9}, "model_id": "qwen/qwen3-vl-235b-a22b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen3 VL 2B Instruct", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "qwen/qwen3-vl-2b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen3 VL 30B A3B Instruct", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "qwen/qwen3-vl-30b-a3b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.2, "display_name": "Qwen3-VL 30B-A3B", "fits": {"amd_rx_7900_xt": 37.3, "amd_rx_7900_xtx": 44.8, "apple_m1_max": 18.7, "apple_m2_max": 18.7, "apple_m2_ultra": 37.3, "apple_m3_max": 18.7, "apple_m3_ultra": 38.2, "apple_m4": 5.6, "apple_m4_max": 25.5, "apple_m4_pro": 12.7, "nvidia_rtx_3090": 43.7, "nvidia_rtx_4090": 47.0, "nvidia_rtx_5090": 83.6}, "model_id": "qwen/qwen3-vl-30b-a3b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"docvqa": 96.5, "mmmu": 72.8, "ocrbench": 87.5}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen3 VL 32B Instruct", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "qwen/qwen3-vl-32b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen3 VL 4B Instruct", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "qwen/qwen3-vl-4b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"ai2d": 85.7, "docvqa": 96.1, "mmmu": 62.5, "ocrbench": 89.6}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen3 VL 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "qwen/qwen3-vl-8b-instruct", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 49.5, "miracl": 51.0, "mteb_classification": 66.0, "mteb_clustering": 44.0, "mteb_overall": 60.0, "mteb_pair_classification": 82.0, "mteb_reranking": 60.0, "mteb_retrieval": 56.5, "mteb_sts": 77.5, "mteb_summarization": 28.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 VL Embedding 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "qwen/qwen3-vl-embedding-2b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 57.5, "miracl": 60.0, "mteb_classification": 72.5, "mteb_clustering": 52.0, "mteb_overall": 68.0, "mteb_pair_classification": 85.5, "mteb_reranking": 65.6, "mteb_retrieval": 65.5, "mteb_sts": 81.1, "mteb_summarization": 30.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Qwen3 VL Embedding 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "qwen/qwen3-vl-embedding-8b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 262144, "cost_input": 0.2, "display_name": "Qwen3-VL Plus", "fits": {}, "model_id": "qwen/qwen3-vl-plus", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": null}, {"benchmark_scores": {"gpqa_diamond": 58.5, "humaneval": 82.5, "ifeval": 80.0, "math_500": 90.2, "mmlu_pro": 72.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.8, "display_name": "QwQ Plus", "fits": {}, "model_id": "qwen/qwq-plus", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "Alibaba / Qwen Team", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 4 14B pile", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "rwkv/rwkv-4-14b-pile", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 4 169M pile", "fits": {"amd_rx_7900_xt": 6627.2, "amd_rx_7900_xtx": 7952.7, "amd_rx_9070_xt": 5343.2, "apple_m1_max": 3313.6, "apple_m2_max": 3313.6, "apple_m2_ultra": 6627.2, "apple_m3_max": 3313.6, "apple_m3_ultra": 6784.6, "apple_m4": 994.1, "apple_m4_max": 4523.1, "apple_m4_pro": 2261.5, "nvidia_rtx_3060_12gb": 2982.2, "nvidia_rtx_3090": 7753.8, "nvidia_rtx_4060_ti_16gb": 2385.8, "nvidia_rtx_4070_ti_super": 5566.9, "nvidia_rtx_4080_super": 6097.0, "nvidia_rtx_4090": 8350.3, "nvidia_rtx_5080": 7952.7, "nvidia_rtx_5090": 14845.0}, "model_id": "rwkv/rwkv-4-169m-pile", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 4 1b5 pile", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "rwkv/rwkv-4-1b5-pile", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 4 3B pile", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "rwkv/rwkv-4-3b-pile", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 4 430M pile", "fits": {"amd_rx_7900_xt": 2604.7, "amd_rx_7900_xtx": 3125.6, "amd_rx_9070_xt": 2100.0, "apple_m1_max": 1302.3, "apple_m2_max": 1302.3, "apple_m2_ultra": 2604.7, "apple_m3_max": 1302.3, "apple_m3_ultra": 2666.5, "apple_m4": 390.7, "apple_m4_max": 1777.7, "apple_m4_pro": 888.8, "nvidia_rtx_3060_12gb": 1172.1, "nvidia_rtx_3090": 3047.4, "nvidia_rtx_4060_ti_16gb": 937.7, "nvidia_rtx_4070_ti_super": 2187.9, "nvidia_rtx_4080_super": 2396.3, "nvidia_rtx_4090": 3281.9, "nvidia_rtx_5080": 3125.6, "nvidia_rtx_5090": 5834.4}, "model_id": "rwkv/rwkv-4-430m-pile", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 4 7B pile", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "rwkv/rwkv-4-7b-pile", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 5 world 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "rwkv/rwkv-5-world-3b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 6 world 1b6", "fits": {"amd_rx_7900_xt": 700.0, "amd_rx_7900_xtx": 840.0, "amd_rx_9070_xt": 564.4, "apple_m1_max": 350.0, "apple_m2_max": 350.0, "apple_m2_ultra": 700.0, "apple_m3_max": 350.0, "apple_m3_ultra": 716.6, "apple_m4": 105.0, "apple_m4_max": 477.7, "apple_m4_pro": 238.9, "nvidia_rtx_3060_12gb": 315.0, "nvidia_rtx_3090": 819.0, "nvidia_rtx_4060_ti_16gb": 252.0, "nvidia_rtx_4070_ti_super": 588.0, "nvidia_rtx_4080_super": 644.0, "nvidia_rtx_4090": 882.0, "nvidia_rtx_5080": 840.0, "nvidia_rtx_5090": 1568.0}, "model_id": "rwkv/rwkv-6-world-1b6", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv 6 world 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "rwkv/rwkv-6-world-7b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {"bbh": 33.1, "gpqa_diamond": 22.9, "ifeval": 7.7, "math_500": 0.5, "mmlu_pro": 11.5, "musr": 39.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv raven 14B", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "rwkv/rwkv-raven-14b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv raven 1b5", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "rwkv/rwkv-raven-1b5", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv raven 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "rwkv/rwkv-raven-3b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "rwkv raven 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "rwkv/rwkv-raven-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "RWKV7 Goose World3 1.5B HF", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "rwkv/rwkv7-goose-world3-1-5b-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "RWKV7 Goose World3 2.9B HF", "fits": {"amd_rx_7900_xt": 386.2, "amd_rx_7900_xtx": 463.4, "amd_rx_9070_xt": 311.4, "apple_m1_max": 193.1, "apple_m2_max": 193.1, "apple_m2_ultra": 386.2, "apple_m3_max": 193.1, "apple_m3_ultra": 395.4, "apple_m4": 57.9, "apple_m4_max": 263.6, "apple_m4_pro": 131.8, "nvidia_rtx_3060_12gb": 173.8, "nvidia_rtx_3090": 451.9, "nvidia_rtx_4060_ti_16gb": 139.0, "nvidia_rtx_4070_ti_super": 324.4, "nvidia_rtx_4080_super": 355.3, "nvidia_rtx_4090": 486.6, "nvidia_rtx_5080": 463.4, "nvidia_rtx_5090": 865.1}, "model_id": "rwkv/rwkv7-goose-world3-2-9b-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "v5 Eagle 7B HF", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "rwkv/v5-eagle-7b-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "v6 Finch 14B HF", "fits": {"amd_rx_7900_xt": 80.0, "amd_rx_7900_xtx": 96.0, "amd_rx_9070_xt": 64.5, "apple_m1_max": 40.0, "apple_m2_max": 40.0, "apple_m2_ultra": 80.0, "apple_m3_max": 40.0, "apple_m3_ultra": 81.9, "apple_m4": 12.0, "apple_m4_max": 54.6, "apple_m4_pro": 27.3, "nvidia_rtx_3060_12gb": 36.0, "nvidia_rtx_3090": 93.6, "nvidia_rtx_4060_ti_16gb": 28.8, "nvidia_rtx_4070_ti_super": 67.2, "nvidia_rtx_4080_super": 73.6, "nvidia_rtx_4090": 100.8, "nvidia_rtx_5080": 96.0, "nvidia_rtx_5090": 179.2}, "model_id": "rwkv/v6-finch-14b-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "v6 Finch 1B6 HF", "fits": {"amd_rx_7900_xt": 700.0, "amd_rx_7900_xtx": 840.0, "amd_rx_9070_xt": 564.4, "apple_m1_max": 350.0, "apple_m2_max": 350.0, "apple_m2_ultra": 700.0, "apple_m3_max": 350.0, "apple_m3_ultra": 716.6, "apple_m4": 105.0, "apple_m4_max": 477.7, "apple_m4_pro": 238.9, "nvidia_rtx_3060_12gb": 315.0, "nvidia_rtx_3090": 819.0, "nvidia_rtx_4060_ti_16gb": 252.0, "nvidia_rtx_4070_ti_super": 588.0, "nvidia_rtx_4080_super": 644.0, "nvidia_rtx_4090": 882.0, "nvidia_rtx_5080": 840.0, "nvidia_rtx_5090": 1568.0}, "model_id": "rwkv/v6-finch-1b6-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "v6 Finch 3B HF", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "rwkv/v6-finch-3b-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "v6 Finch 7B HF", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "rwkv/v6-finch-7b-hf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "RWKV Foundation", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "blip image captioning base", "fits": {"amd_rx_7900_xt": 5000.0, "amd_rx_7900_xtx": 6000.0, "amd_rx_9070_xt": 4031.2, "apple_m1_max": 2500.0, "apple_m2_max": 2500.0, "apple_m2_ultra": 5000.0, "apple_m3_max": 2500.0, "apple_m3_ultra": 5118.7, "apple_m4": 750.0, "apple_m4_max": 3412.5, "apple_m4_pro": 1706.2, "nvidia_rtx_3060_12gb": 2250.0, "nvidia_rtx_3090": 5850.0, "nvidia_rtx_4060_ti_16gb": 1800.0, "nvidia_rtx_4070_ti_super": 4200.0, "nvidia_rtx_4080_super": 4600.0, "nvidia_rtx_4090": 6300.0, "nvidia_rtx_5080": 6000.0, "nvidia_rtx_5090": 11200.0}, "model_id": "salesforce/blip-image-captioning-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "blip image captioning large", "fits": {"amd_rx_7900_xt": 2383.0, "amd_rx_7900_xtx": 2859.6, "amd_rx_9070_xt": 1921.3, "apple_m1_max": 1191.5, "apple_m2_max": 1191.5, "apple_m2_ultra": 2383.0, "apple_m3_max": 1191.5, "apple_m3_ultra": 2439.6, "apple_m4": 357.4, "apple_m4_max": 1626.4, "apple_m4_pro": 813.2, "nvidia_rtx_3060_12gb": 1072.3, "nvidia_rtx_3090": 2788.1, "nvidia_rtx_4060_ti_16gb": 857.9, "nvidia_rtx_4070_ti_super": 2001.7, "nvidia_rtx_4080_super": 2192.3, "nvidia_rtx_4090": 3002.6, "nvidia_rtx_5080": 2859.6, "nvidia_rtx_5090": 5337.9}, "model_id": "salesforce/blip-image-captioning-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "blip itm base coco", "fits": {"amd_rx_7900_xt": 5000.0, "amd_rx_7900_xtx": 6000.0, "amd_rx_9070_xt": 4031.2, "apple_m1_max": 2500.0, "apple_m2_max": 2500.0, "apple_m2_ultra": 5000.0, "apple_m3_max": 2500.0, "apple_m3_ultra": 5118.7, "apple_m4": 750.0, "apple_m4_max": 3412.5, "apple_m4_pro": 1706.2, "nvidia_rtx_3060_12gb": 2250.0, "nvidia_rtx_3090": 5850.0, "nvidia_rtx_4060_ti_16gb": 1800.0, "nvidia_rtx_4070_ti_super": 4200.0, "nvidia_rtx_4080_super": 4600.0, "nvidia_rtx_4090": 6300.0, "nvidia_rtx_5080": 6000.0, "nvidia_rtx_5090": 11200.0}, "model_id": "salesforce/blip-itm-base-coco", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 512, "cost_input": null, "display_name": "blip vqa base", "fits": {"amd_rx_7900_xt": 5000.0, "amd_rx_7900_xtx": 6000.0, "amd_rx_9070_xt": 4031.2, "apple_m1_max": 2500.0, "apple_m2_max": 2500.0, "apple_m2_ultra": 5000.0, "apple_m3_max": 2500.0, "apple_m3_ultra": 5118.7, "apple_m4": 750.0, "apple_m4_max": 3412.5, "apple_m4_pro": 1706.2, "nvidia_rtx_3060_12gb": 2250.0, "nvidia_rtx_3090": 5850.0, "nvidia_rtx_4060_ti_16gb": 1800.0, "nvidia_rtx_4070_ti_super": 4200.0, "nvidia_rtx_4080_super": 4600.0, "nvidia_rtx_4090": 6300.0, "nvidia_rtx_5080": 6000.0, "nvidia_rtx_5090": 11200.0}, "model_id": "salesforce/blip-vqa-base", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 512, "cost_input": null, "display_name": "blip vqa capfilt large", "fits": {"amd_rx_7900_xt": 2383.0, "amd_rx_7900_xtx": 2859.6, "amd_rx_9070_xt": 1921.3, "apple_m1_max": 1191.5, "apple_m2_max": 1191.5, "apple_m2_ultra": 2383.0, "apple_m3_max": 1191.5, "apple_m3_ultra": 2439.6, "apple_m4": 357.4, "apple_m4_max": 1626.4, "apple_m4_pro": 813.2, "nvidia_rtx_3060_12gb": 1072.3, "nvidia_rtx_3090": 2788.1, "nvidia_rtx_4060_ti_16gb": 857.9, "nvidia_rtx_4070_ti_super": 2001.7, "nvidia_rtx_4080_super": 2192.3, "nvidia_rtx_4090": 3002.6, "nvidia_rtx_5080": 2859.6, "nvidia_rtx_5090": 5337.9}, "model_id": "salesforce/blip-vqa-capfilt-large", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 512, "cost_input": null, "display_name": "blip2 flan t5 xl", "fits": {"amd_rx_7900_xt": 273.2, "amd_rx_7900_xtx": 327.8, "amd_rx_9070_xt": 220.2, "apple_m1_max": 136.6, "apple_m2_max": 136.6, "apple_m2_ultra": 273.2, "apple_m3_max": 136.6, "apple_m3_ultra": 279.7, "apple_m4": 41.0, "apple_m4_max": 186.4, "apple_m4_pro": 93.2, "nvidia_rtx_3060_12gb": 122.9, "nvidia_rtx_3090": 319.6, "nvidia_rtx_4060_ti_16gb": 98.3, "nvidia_rtx_4070_ti_super": 229.5, "nvidia_rtx_4080_super": 251.3, "nvidia_rtx_4090": 344.2, "nvidia_rtx_5080": 327.8, "nvidia_rtx_5090": 611.9}, "model_id": "salesforce/blip2-flan-t5-xl", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "blip2 itm vit g", "fits": {"amd_rx_7900_xt": 933.3, "amd_rx_7900_xtx": 1120.0, "amd_rx_9070_xt": 752.5, "apple_m1_max": 466.7, "apple_m2_max": 466.7, "apple_m2_ultra": 933.3, "apple_m3_max": 466.7, "apple_m3_ultra": 955.5, "apple_m4": 140.0, "apple_m4_max": 637.0, "apple_m4_pro": 318.5, "nvidia_rtx_3060_12gb": 420.0, "nvidia_rtx_3090": 1092.0, "nvidia_rtx_4060_ti_16gb": 336.0, "nvidia_rtx_4070_ti_super": 784.0, "nvidia_rtx_4080_super": 858.7, "nvidia_rtx_4090": 1176.0, "nvidia_rtx_5080": 1120.0, "nvidia_rtx_5090": 2090.7}, "model_id": "salesforce/blip2-itm-vit-g", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "blip2 opt 2.7B coco", "fits": {"amd_rx_7900_xt": 414.8, "amd_rx_7900_xtx": 497.8, "amd_rx_9070_xt": 334.4, "apple_m1_max": 207.4, "apple_m2_max": 207.4, "apple_m2_ultra": 414.8, "apple_m3_max": 207.4, "apple_m3_ultra": 424.7, "apple_m4": 62.2, "apple_m4_max": 283.1, "apple_m4_pro": 141.6, "nvidia_rtx_3060_12gb": 186.7, "nvidia_rtx_3090": 485.3, "nvidia_rtx_4060_ti_16gb": 149.3, "nvidia_rtx_4070_ti_super": 348.4, "nvidia_rtx_4080_super": 381.6, "nvidia_rtx_4090": 522.7, "nvidia_rtx_5080": 497.8, "nvidia_rtx_5090": 929.2}, "model_id": "salesforce/blip2-opt-2-7b-coco", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 512, "cost_input": null, "display_name": "blip2 opt 2.7B", "fits": {"amd_rx_7900_xt": 414.8, "amd_rx_7900_xtx": 497.8, "amd_rx_9070_xt": 334.4, "apple_m1_max": 207.4, "apple_m2_max": 207.4, "apple_m2_ultra": 414.8, "apple_m3_max": 207.4, "apple_m3_ultra": 424.7, "apple_m4": 62.2, "apple_m4_max": 283.1, "apple_m4_pro": 141.6, "nvidia_rtx_3060_12gb": 186.7, "nvidia_rtx_3090": 485.3, "nvidia_rtx_4060_ti_16gb": 149.3, "nvidia_rtx_4070_ti_super": 348.4, "nvidia_rtx_4080_super": 381.6, "nvidia_rtx_4090": 522.7, "nvidia_rtx_5080": 497.8, "nvidia_rtx_5090": 929.2}, "model_id": "salesforce/blip2-opt-2-7b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 512, "cost_input": null, "display_name": "blip2 opt 6.7B", "fits": {"amd_rx_7900_xt": 167.2, "amd_rx_7900_xtx": 200.6, "amd_rx_9070_xt": 134.8, "apple_m1_max": 83.6, "apple_m2_max": 83.6, "apple_m2_ultra": 167.2, "apple_m3_max": 83.6, "apple_m3_ultra": 171.1, "apple_m4": 25.1, "apple_m4_max": 114.1, "apple_m4_pro": 57.0, "nvidia_rtx_3060_12gb": 75.2, "nvidia_rtx_3090": 195.6, "nvidia_rtx_4060_ti_16gb": 60.2, "nvidia_rtx_4070_ti_super": 140.4, "nvidia_rtx_4080_super": 153.8, "nvidia_rtx_4090": 210.6, "nvidia_rtx_5080": 200.6, "nvidia_rtx_5090": 374.4}, "model_id": "salesforce/blip2-opt-6-7b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "codegen 350M mono", "fits": {"amd_rx_7900_xt": 3200.0, "amd_rx_7900_xtx": 3840.0, "amd_rx_9070_xt": 2580.0, "apple_m1_max": 1600.0, "apple_m2_max": 1600.0, "apple_m2_ultra": 3200.0, "apple_m3_max": 1600.0, "apple_m3_ultra": 3276.0, "apple_m4": 480.0, "apple_m4_max": 2184.0, "apple_m4_pro": 1092.0, "nvidia_rtx_3060_12gb": 1440.0, "nvidia_rtx_3090": 3744.0, "nvidia_rtx_4060_ti_16gb": 1152.0, "nvidia_rtx_4070_ti_super": 2688.0, "nvidia_rtx_4080_super": 2944.0, "nvidia_rtx_4090": 4032.0, "nvidia_rtx_5080": 3840.0, "nvidia_rtx_5090": 7168.0}, "model_id": "salesforce/codegen-350m-mono", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 512, "cost_input": null, "display_name": "codet5 small", "fits": {"amd_rx_7900_xt": 18666.7, "amd_rx_7900_xtx": 22400.0, "amd_rx_9070_xt": 15050.0, "apple_m1_max": 9333.3, "apple_m2_max": 9333.3, "apple_m2_ultra": 18666.7, "apple_m3_max": 9333.3, "apple_m3_ultra": 19110.0, "apple_m4": 2800.0, "apple_m4_max": 12740.0, "apple_m4_pro": 6370.0, "nvidia_rtx_3060_12gb": 8400.0, "nvidia_rtx_3090": 21840.0, "nvidia_rtx_4060_ti_16gb": 6720.0, "nvidia_rtx_4070_ti_super": 15680.0, "nvidia_rtx_4080_super": 17173.3, "nvidia_rtx_4090": 23520.0, "nvidia_rtx_5080": 22400.0, "nvidia_rtx_5090": 41813.3}, "model_id": "salesforce/codet5-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "codet5p 110M embedding", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "salesforce/codet5p-110m-embedding", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "codet5p 220M", "fits": {"amd_rx_7900_xt": 5090.9, "amd_rx_7900_xtx": 6109.1, "amd_rx_9070_xt": 4104.5, "apple_m1_max": 2545.5, "apple_m2_max": 2545.5, "apple_m2_ultra": 5090.9, "apple_m3_max": 2545.5, "apple_m3_ultra": 5211.8, "apple_m4": 763.6, "apple_m4_max": 3474.5, "apple_m4_pro": 1737.3, "nvidia_rtx_3060_12gb": 2290.9, "nvidia_rtx_3090": 5956.4, "nvidia_rtx_4060_ti_16gb": 1832.7, "nvidia_rtx_4070_ti_super": 4276.4, "nvidia_rtx_4080_super": 4683.6, "nvidia_rtx_4090": 6414.5, "nvidia_rtx_5080": 6109.1, "nvidia_rtx_5090": 11403.6}, "model_id": "salesforce/codet5p-220m", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256, "cost_input": null, "display_name": "ctrl", "fits": {"amd_rx_7900_xt": 687.1, "amd_rx_7900_xtx": 824.5, "amd_rx_9070_xt": 554.0, "apple_m1_max": 343.6, "apple_m2_max": 343.6, "apple_m2_ultra": 687.1, "apple_m3_max": 343.6, "apple_m3_ultra": 703.4, "apple_m4": 103.1, "apple_m4_max": 469.0, "apple_m4_pro": 234.5, "nvidia_rtx_3060_12gb": 309.2, "nvidia_rtx_3090": 803.9, "nvidia_rtx_4060_ti_16gb": 247.4, "nvidia_rtx_4070_ti_super": 577.2, "nvidia_rtx_4080_super": 632.1, "nvidia_rtx_4090": 865.8, "nvidia_rtx_5080": 824.5, "nvidia_rtx_5090": 1539.1}, "model_id": "salesforce/ctrl", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Llama xLAM 2 8B fc r", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "salesforce/llama-xlam-2-8b-fc-r", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "moirai 1.0 R large", "fits": {"amd_rx_7900_xt": 3601.3, "amd_rx_7900_xtx": 4321.5, "amd_rx_9070_xt": 2903.5, "apple_m1_max": 1800.6, "apple_m2_max": 1800.6, "apple_m2_ultra": 3601.3, "apple_m3_max": 1800.6, "apple_m3_ultra": 3686.8, "apple_m4": 540.2, "apple_m4_max": 2457.9, "apple_m4_pro": 1228.9, "nvidia_rtx_3060_12gb": 1620.6, "nvidia_rtx_3090": 4213.5, "nvidia_rtx_4060_ti_16gb": 1296.5, "nvidia_rtx_4070_ti_super": 3025.1, "nvidia_rtx_4080_super": 3313.2, "nvidia_rtx_4090": 4537.6, "nvidia_rtx_5080": 4321.5, "nvidia_rtx_5090": 8066.9}, "model_id": "salesforce/moirai-1-0-r-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "moirai 1.1 R base", "fits": {"amd_rx_7900_xt": 12307.7, "amd_rx_7900_xtx": 14769.2, "amd_rx_9070_xt": 9923.1, "apple_m1_max": 6153.8, "apple_m2_max": 6153.8, "apple_m2_ultra": 12307.7, "apple_m3_max": 6153.8, "apple_m3_ultra": 12600.0, "apple_m4": 1846.2, "apple_m4_max": 8400.0, "apple_m4_pro": 4200.0, "nvidia_rtx_3060_12gb": 5538.5, "nvidia_rtx_3090": 14400.0, "nvidia_rtx_4060_ti_16gb": 4430.8, "nvidia_rtx_4070_ti_super": 10338.5, "nvidia_rtx_4080_super": 11323.1, "nvidia_rtx_4090": 15507.7, "nvidia_rtx_5080": 14769.2, "nvidia_rtx_5090": 27569.2}, "model_id": "salesforce/moirai-1-1-r-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "moirai 1.1 R large", "fits": {"amd_rx_7900_xt": 3601.3, "amd_rx_7900_xtx": 4321.5, "amd_rx_9070_xt": 2903.5, "apple_m1_max": 1800.6, "apple_m2_max": 1800.6, "apple_m2_ultra": 3601.3, "apple_m3_max": 1800.6, "apple_m3_ultra": 3686.8, "apple_m4": 540.2, "apple_m4_max": 2457.9, "apple_m4_pro": 1228.9, "nvidia_rtx_3060_12gb": 1620.6, "nvidia_rtx_3090": 4213.5, "nvidia_rtx_4060_ti_16gb": 1296.5, "nvidia_rtx_4070_ti_super": 3025.1, "nvidia_rtx_4080_super": 3313.2, "nvidia_rtx_4090": 4537.6, "nvidia_rtx_5080": 4321.5, "nvidia_rtx_5090": 8066.9}, "model_id": "salesforce/moirai-1-1-r-large", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "moirai 1.1 R small", "fits": {"amd_rx_7900_xt": 80000.0, "amd_rx_7900_xtx": 96000.0, "amd_rx_9070_xt": 64500.0, "apple_m1_max": 40000.0, "apple_m2_max": 40000.0, "apple_m2_ultra": 80000.0, "apple_m3_max": 40000.0, "apple_m3_ultra": 81900.0, "apple_m4": 12000.0, "apple_m4_max": 54600.0, "apple_m4_pro": 27300.0, "nvidia_rtx_3060_12gb": 36000.0, "nvidia_rtx_3090": 93600.0, "nvidia_rtx_4060_ti_16gb": 28800.0, "nvidia_rtx_4070_ti_super": 67200.0, "nvidia_rtx_4080_super": 73600.0, "nvidia_rtx_4090": 100800.0, "nvidia_rtx_5080": 96000.0, "nvidia_rtx_5090": 179200.0}, "model_id": "salesforce/moirai-1-1-r-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "moirai 2.0 R small", "fits": {"amd_rx_7900_xt": 80000.0, "amd_rx_7900_xtx": 96000.0, "amd_rx_9070_xt": 64500.0, "apple_m1_max": 40000.0, "apple_m2_max": 40000.0, "apple_m2_ultra": 80000.0, "apple_m3_max": 40000.0, "apple_m3_ultra": 81900.0, "apple_m4": 12000.0, "apple_m4_max": 54600.0, "apple_m4_pro": 27300.0, "nvidia_rtx_3060_12gb": 36000.0, "nvidia_rtx_3090": 93600.0, "nvidia_rtx_4060_ti_16gb": 28800.0, "nvidia_rtx_4070_ti_super": 67200.0, "nvidia_rtx_4080_super": 73600.0, "nvidia_rtx_4090": 100800.0, "nvidia_rtx_5080": 96000.0, "nvidia_rtx_5090": 179200.0}, "model_id": "salesforce/moirai-2-0-r-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "moirai moe 1.0 R base", "fits": {"amd_rx_7900_xt": 12307.7, "amd_rx_7900_xtx": 14769.2, "amd_rx_9070_xt": 9923.1, "apple_m1_max": 6153.8, "apple_m2_max": 6153.8, "apple_m2_ultra": 12307.7, "apple_m3_max": 6153.8, "apple_m3_ultra": 12600.0, "apple_m4": 1846.2, "apple_m4_max": 8400.0, "apple_m4_pro": 4200.0, "nvidia_rtx_3060_12gb": 5538.5, "nvidia_rtx_3090": 14400.0, "nvidia_rtx_4060_ti_16gb": 4430.8, "nvidia_rtx_4070_ti_super": 10338.5, "nvidia_rtx_4080_super": 11323.1, "nvidia_rtx_4090": 15507.7, "nvidia_rtx_5080": 14769.2, "nvidia_rtx_5090": 27569.2}, "model_id": "salesforce/moirai-moe-1-0-r-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "moirai moe 1.0 R small", "fits": {"amd_rx_7900_xt": 80000.0, "amd_rx_7900_xtx": 96000.0, "amd_rx_9070_xt": 64500.0, "apple_m1_max": 40000.0, "apple_m2_max": 40000.0, "apple_m2_ultra": 80000.0, "apple_m3_max": 40000.0, "apple_m3_ultra": 81900.0, "apple_m4": 12000.0, "apple_m4_max": 54600.0, "apple_m4_pro": 27300.0, "nvidia_rtx_3060_12gb": 36000.0, "nvidia_rtx_3090": 93600.0, "nvidia_rtx_4060_ti_16gb": 28800.0, "nvidia_rtx_4070_ti_super": 67200.0, "nvidia_rtx_4080_super": 73600.0, "nvidia_rtx_4090": 100800.0, "nvidia_rtx_5080": 96000.0, "nvidia_rtx_5090": 179200.0}, "model_id": "salesforce/moirai-moe-1-0-r-small", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {"beir": 55.0, "miracl": 50.0, "mteb_classification": 70.8, "mteb_clustering": 49.0, "mteb_overall": 65.2, "mteb_pair_classification": 86.0, "mteb_reranking": 58.5, "mteb_retrieval": 62.5, "mteb_sts": 84.0, "mteb_summarization": 30.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "SFR Embedding 2 R", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "salesforce/sfr-embedding-2-r", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Salesforce", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 35.0, "mteb_classification": 65.2, "mteb_clustering": 43.2, "mteb_overall": 59.5, "mteb_pair_classification": 82.5, "mteb_reranking": 54.0, "mteb_retrieval": 55.8, "mteb_sts": 79.5, "mteb_summarization": 28.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "SFR Embedding Code 400M R", "fits": {"amd_rx_7900_xt": 2800.0, "amd_rx_7900_xtx": 3360.0, "amd_rx_9070_xt": 2257.5, "apple_m1_max": 1400.0, "apple_m2_max": 1400.0, "apple_m2_ultra": 2800.0, "apple_m3_max": 1400.0, "apple_m3_ultra": 2866.5, "apple_m4": 420.0, "apple_m4_max": 1911.0, "apple_m4_pro": 955.5, "nvidia_rtx_3060_12gb": 1260.0, "nvidia_rtx_3090": 3276.0, "nvidia_rtx_4060_ti_16gb": 1008.0, "nvidia_rtx_4070_ti_super": 2352.0, "nvidia_rtx_4080_super": 2576.0, "nvidia_rtx_4090": 3528.0, "nvidia_rtx_5080": 3360.0, "nvidia_rtx_5090": 6272.0}, "model_id": "salesforce/sfr-embedding-code-400m-r", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Salesforce", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "xLAM 2 1B fc r", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "salesforce/xlam-2-1b-fc-r", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "xLAM 2 32B fc r", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "salesforce/xlam-2-32b-fc-r", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "xLAM 2 3B fc r", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "salesforce/xlam-2-3b-fc-r", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "xLAM 7B r", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "salesforce/xlam-7b-r", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "xLAM 8x7b r", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "salesforce/xlam-8x7b-r", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Salesforce", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "SGuard ContentFilter 2B v1", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "samsung/sguard-contentfilter-2b-v1", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Samsung", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "SGuard JailbreakFilter 2B v1", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "samsung/sguard-jailbreakfilter-2b-v1", "model_subtypes": [], "model_type": "safety-classifier", "open_weights": true, "provider": "Samsung", "scores_as_of": null}, {"benchmark_scores": {"beir": 41.5, "miracl": 27.0, "mteb_classification": 62.0, "mteb_clustering": 39.0, "mteb_overall": 55.2, "mteb_pair_classification": 79.0, "mteb_reranking": 48.5, "mteb_retrieval": 47.5, "mteb_sts": 76.5, "mteb_summarization": 25.8}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "all distilroberta v1", "fits": {"amd_rx_7900_xt": 13658.5, "amd_rx_7900_xtx": 16390.2, "amd_rx_9070_xt": 11012.2, "apple_m1_max": 6829.3, "apple_m2_max": 6829.3, "apple_m2_ultra": 13658.5, "apple_m3_max": 6829.3, "apple_m3_ultra": 13982.9, "apple_m4": 2048.8, "apple_m4_max": 9322.0, "apple_m4_pro": 4661.0, "nvidia_rtx_3060_12gb": 6146.3, "nvidia_rtx_3090": 15980.5, "nvidia_rtx_4060_ti_16gb": 4917.1, "nvidia_rtx_4070_ti_super": 11473.2, "nvidia_rtx_4080_super": 12565.9, "nvidia_rtx_4090": 17209.8, "nvidia_rtx_5080": 16390.2, "nvidia_rtx_5090": 30595.1}, "model_id": "sentence-transformers/all-distilroberta-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 43.2, "miracl": 29.0, "mteb_classification": 63.8, "mteb_clustering": 40.5, "mteb_overall": 56.8, "mteb_pair_classification": 81.0, "mteb_reranking": 50.0, "mteb_retrieval": 49.2, "mteb_sts": 78.8, "mteb_summarization": 26.8}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "all MiniLM L12 v2", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "sentence-transformers/all-minilm-l12-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 42.8, "miracl": 28.0, "mteb_classification": 63.2, "mteb_clustering": 40.1, "mteb_overall": 56.3, "mteb_pair_classification": 80.5, "mteb_reranking": 49.5, "mteb_retrieval": 48.5, "mteb_sts": 78.2, "mteb_summarization": 26.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "all MiniLM L6 v2", "fits": {"amd_rx_7900_xt": 49339.2, "amd_rx_7900_xtx": 59207.0, "amd_rx_9070_xt": 39779.7, "apple_m1_max": 24669.6, "apple_m2_max": 24669.6, "apple_m2_ultra": 49339.2, "apple_m3_max": 24669.6, "apple_m3_ultra": 50511.0, "apple_m4": 7400.9, "apple_m4_max": 33674.0, "apple_m4_pro": 16837.0, "nvidia_rtx_3060_12gb": 22202.6, "nvidia_rtx_3090": 57726.9, "nvidia_rtx_4060_ti_16gb": 17762.1, "nvidia_rtx_4070_ti_super": 41444.9, "nvidia_rtx_4080_super": 45392.1, "nvidia_rtx_4090": 62167.4, "nvidia_rtx_5080": 59207.0, "nvidia_rtx_5090": 110519.8}, "model_id": "sentence-transformers/all-minilm-l6-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 44.2, "miracl": 30.0, "mteb_classification": 64.5, "mteb_clustering": 41.5, "mteb_overall": 57.8, "mteb_pair_classification": 81.5, "mteb_reranking": 50.8, "mteb_retrieval": 50.2, "mteb_sts": 79.5, "mteb_summarization": 27.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "all mpnet base v2", "fits": {"amd_rx_7900_xt": 10275.2, "amd_rx_7900_xtx": 12330.3, "amd_rx_9070_xt": 8284.4, "apple_m1_max": 5137.6, "apple_m2_max": 5137.6, "apple_m2_ultra": 10275.2, "apple_m3_max": 5137.6, "apple_m3_ultra": 10519.3, "apple_m4": 1541.3, "apple_m4_max": 7012.8, "apple_m4_pro": 3506.4, "nvidia_rtx_3060_12gb": 4623.9, "nvidia_rtx_3090": 12022.0, "nvidia_rtx_4060_ti_16gb": 3699.1, "nvidia_rtx_4070_ti_super": 8631.2, "nvidia_rtx_4080_super": 9453.2, "nvidia_rtx_4090": 12946.8, "nvidia_rtx_5080": 12330.3, "nvidia_rtx_5090": 23016.5}, "model_id": "sentence-transformers/all-mpnet-base-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 42.0, "miracl": 28.5, "mteb_classification": 62.8, "mteb_clustering": 39.5, "mteb_overall": 56.0, "mteb_pair_classification": 79.5, "mteb_reranking": 49.0, "mteb_retrieval": 48.0, "mteb_sts": 77.2, "mteb_summarization": 26.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "all roberta large v1", "fits": {"amd_rx_7900_xt": 3154.9, "amd_rx_7900_xtx": 3785.9, "amd_rx_9070_xt": 2543.7, "apple_m1_max": 1577.5, "apple_m2_max": 1577.5, "apple_m2_ultra": 3154.9, "apple_m3_max": 1577.5, "apple_m3_ultra": 3229.9, "apple_m4": 473.2, "apple_m4_max": 2153.2, "apple_m4_pro": 1076.6, "nvidia_rtx_3060_12gb": 1419.7, "nvidia_rtx_3090": 3691.3, "nvidia_rtx_4060_ti_16gb": 1135.8, "nvidia_rtx_4070_ti_super": 2650.1, "nvidia_rtx_4080_super": 2902.5, "nvidia_rtx_4090": 3975.2, "nvidia_rtx_5080": 3785.9, "nvidia_rtx_5090": 7067.0}, "model_id": "sentence-transformers/all-roberta-large-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 36.0, "mteb_classification": 55.5, "mteb_clustering": 32.5, "mteb_overall": 48.5, "mteb_pair_classification": 73.5, "mteb_reranking": 43.0, "mteb_retrieval": 38.8, "mteb_sts": 70.5, "mteb_summarization": 22.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "distiluse base multilingual cased v1", "fits": {"amd_rx_7900_xt": 8296.3, "amd_rx_7900_xtx": 9955.6, "amd_rx_9070_xt": 6688.9, "apple_m1_max": 4148.1, "apple_m2_max": 4148.1, "apple_m2_ultra": 8296.3, "apple_m3_max": 4148.1, "apple_m3_ultra": 8493.3, "apple_m4": 1244.4, "apple_m4_max": 5662.2, "apple_m4_pro": 2831.1, "nvidia_rtx_3060_12gb": 3733.3, "nvidia_rtx_3090": 9706.7, "nvidia_rtx_4060_ti_16gb": 2986.7, "nvidia_rtx_4070_ti_super": 6968.9, "nvidia_rtx_4080_super": 7632.6, "nvidia_rtx_4090": 10453.3, "nvidia_rtx_5080": 9955.6, "nvidia_rtx_5090": 18583.7}, "model_id": "sentence-transformers/distiluse-base-multilingual-cased-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 38.0, "mteb_classification": 57.0, "mteb_clustering": 34.0, "mteb_overall": 50.0, "mteb_pair_classification": 75.0, "mteb_reranking": 44.0, "mteb_retrieval": 40.5, "mteb_sts": 72.0, "mteb_summarization": 23.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "distiluse base multilingual cased v2", "fits": {"amd_rx_7900_xt": 8296.3, "amd_rx_7900_xtx": 9955.6, "amd_rx_9070_xt": 6688.9, "apple_m1_max": 4148.1, "apple_m2_max": 4148.1, "apple_m2_ultra": 8296.3, "apple_m3_max": 4148.1, "apple_m3_ultra": 8493.3, "apple_m4": 1244.4, "apple_m4_max": 5662.2, "apple_m4_pro": 2831.1, "nvidia_rtx_3060_12gb": 3733.3, "nvidia_rtx_3090": 9706.7, "nvidia_rtx_4060_ti_16gb": 2986.7, "nvidia_rtx_4070_ti_super": 6968.9, "nvidia_rtx_4080_super": 7632.6, "nvidia_rtx_4090": 10453.3, "nvidia_rtx_5080": 9955.6, "nvidia_rtx_5090": 18583.7}, "model_id": "sentence-transformers/distiluse-base-multilingual-cased-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 45.0, "mteb_classification": 59.2, "mteb_clustering": 36.2, "mteb_overall": 52.0, "mteb_pair_classification": 76.0, "mteb_reranking": 45.0, "mteb_retrieval": 43.5, "mteb_sts": 73.0, "mteb_summarization": 24.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "LaBSE", "fits": {"amd_rx_7900_xt": 2377.9, "amd_rx_7900_xtx": 2853.5, "amd_rx_9070_xt": 1917.2, "apple_m1_max": 1189.0, "apple_m2_max": 1189.0, "apple_m2_ultra": 2377.9, "apple_m3_max": 1189.0, "apple_m3_ultra": 2434.4, "apple_m4": 356.7, "apple_m4_max": 1622.9, "apple_m4_pro": 811.5, "nvidia_rtx_3060_12gb": 1070.1, "nvidia_rtx_3090": 2782.2, "nvidia_rtx_4060_ti_16gb": 856.1, "nvidia_rtx_4070_ti_super": 1997.5, "nvidia_rtx_4080_super": 2187.7, "nvidia_rtx_4090": 2996.2, "nvidia_rtx_5080": 2853.5, "nvidia_rtx_5090": 5326.5}, "model_id": "sentence-transformers/labse", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 40.5, "miracl": 29.0, "mteb_classification": 59.5, "mteb_clustering": 36.5, "mteb_overall": 53.0, "mteb_pair_classification": 78.5, "mteb_reranking": 47.5, "mteb_retrieval": 46.5, "mteb_sts": 75.5, "mteb_summarization": 25.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "msmarco bert base dot v5", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "sentence-transformers/msmarco-bert-base-dot-v5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 40.0, "miracl": 28.0, "mteb_classification": 59.0, "mteb_clustering": 36.0, "mteb_overall": 52.5, "mteb_pair_classification": 77.5, "mteb_reranking": 46.5, "mteb_retrieval": 45.5, "mteb_sts": 74.5, "mteb_summarization": 25.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "msmarco MiniLM L12 cos v5", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "sentence-transformers/msmarco-minilm-l12-cos-v5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 39.0, "miracl": 27.5, "mteb_classification": 58.0, "mteb_clustering": 35.0, "mteb_overall": 51.5, "mteb_pair_classification": 77.0, "mteb_reranking": 46.0, "mteb_retrieval": 44.2, "mteb_sts": 74.0, "mteb_summarization": 24.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "msmarco MiniLM L6 v3", "fits": {"amd_rx_7900_xt": 49339.2, "amd_rx_7900_xtx": 59207.0, "amd_rx_9070_xt": 39779.7, "apple_m1_max": 24669.6, "apple_m2_max": 24669.6, "apple_m2_ultra": 49339.2, "apple_m3_max": 24669.6, "apple_m3_ultra": 50511.0, "apple_m4": 7400.9, "apple_m4_max": 33674.0, "apple_m4_pro": 16837.0, "nvidia_rtx_3060_12gb": 22202.6, "nvidia_rtx_3090": 57726.9, "nvidia_rtx_4060_ti_16gb": 17762.1, "nvidia_rtx_4070_ti_super": 41444.9, "nvidia_rtx_4080_super": 45392.1, "nvidia_rtx_4090": 62167.4, "nvidia_rtx_5080": 59207.0, "nvidia_rtx_5090": 110519.8}, "model_id": "sentence-transformers/msmarco-minilm-l6-v3", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 41.0, "miracl": 30.0, "mteb_classification": 61.0, "mteb_clustering": 38.0, "mteb_overall": 54.2, "mteb_pair_classification": 78.5, "mteb_reranking": 48.0, "mteb_retrieval": 46.8, "mteb_sts": 76.0, "mteb_summarization": 25.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "multi qa MiniLM L6 cos v1", "fits": {"amd_rx_7900_xt": 49339.2, "amd_rx_7900_xtx": 59207.0, "amd_rx_9070_xt": 39779.7, "apple_m1_max": 24669.6, "apple_m2_max": 24669.6, "apple_m2_ultra": 49339.2, "apple_m3_max": 24669.6, "apple_m3_ultra": 50511.0, "apple_m4": 7400.9, "apple_m4_max": 33674.0, "apple_m4_pro": 16837.0, "nvidia_rtx_3060_12gb": 22202.6, "nvidia_rtx_3090": 57726.9, "nvidia_rtx_4060_ti_16gb": 17762.1, "nvidia_rtx_4070_ti_super": 41444.9, "nvidia_rtx_4080_super": 45392.1, "nvidia_rtx_4090": 62167.4, "nvidia_rtx_5080": 59207.0, "nvidia_rtx_5090": 110519.8}, "model_id": "sentence-transformers/multi-qa-minilm-l6-cos-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 42.8, "miracl": 31.5, "mteb_classification": 62.2, "mteb_clustering": 39.5, "mteb_overall": 55.5, "mteb_pair_classification": 79.8, "mteb_reranking": 49.2, "mteb_retrieval": 48.2, "mteb_sts": 77.2, "mteb_summarization": 26.2}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "multi qa mpnet base cos v1", "fits": {"amd_rx_7900_xt": 10275.2, "amd_rx_7900_xtx": 12330.3, "amd_rx_9070_xt": 8284.4, "apple_m1_max": 5137.6, "apple_m2_max": 5137.6, "apple_m2_ultra": 10275.2, "apple_m3_max": 5137.6, "apple_m3_ultra": 10519.3, "apple_m4": 1541.3, "apple_m4_max": 7012.8, "apple_m4_pro": 3506.4, "nvidia_rtx_3060_12gb": 4623.9, "nvidia_rtx_3090": 12022.0, "nvidia_rtx_4060_ti_16gb": 3699.1, "nvidia_rtx_4070_ti_super": 8631.2, "nvidia_rtx_4080_super": 9453.2, "nvidia_rtx_4090": 12946.8, "nvidia_rtx_5080": 12330.3, "nvidia_rtx_5090": 23016.5}, "model_id": "sentence-transformers/multi-qa-mpnet-base-cos-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 43.0, "miracl": 32.0, "mteb_classification": 62.5, "mteb_clustering": 39.8, "mteb_overall": 55.8, "mteb_pair_classification": 80.0, "mteb_reranking": 49.5, "mteb_retrieval": 48.5, "mteb_sts": 77.5, "mteb_summarization": 26.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "multi qa mpnet base dot v1", "fits": {"amd_rx_7900_xt": 10275.2, "amd_rx_7900_xtx": 12330.3, "amd_rx_9070_xt": 8284.4, "apple_m1_max": 5137.6, "apple_m2_max": 5137.6, "apple_m2_ultra": 10275.2, "apple_m3_max": 5137.6, "apple_m3_ultra": 10519.3, "apple_m4": 1541.3, "apple_m4_max": 7012.8, "apple_m4_pro": 3506.4, "nvidia_rtx_3060_12gb": 4623.9, "nvidia_rtx_3090": 12022.0, "nvidia_rtx_4060_ti_16gb": 3699.1, "nvidia_rtx_4070_ti_super": 8631.2, "nvidia_rtx_4080_super": 9453.2, "nvidia_rtx_4090": 12946.8, "nvidia_rtx_5080": 12330.3, "nvidia_rtx_5090": 23016.5}, "model_id": "sentence-transformers/multi-qa-mpnet-base-dot-v1", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 27.0, "mteb_classification": 60.2, "mteb_clustering": 37.2, "mteb_overall": 53.2, "mteb_pair_classification": 78.0, "mteb_reranking": 46.8, "mteb_retrieval": 44.5, "mteb_sts": 75.2, "mteb_summarization": 25.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "paraphrase MiniLM L12 v2", "fits": {"amd_rx_7900_xt": 33532.9, "amd_rx_7900_xtx": 40239.5, "amd_rx_9070_xt": 27035.9, "apple_m1_max": 16766.5, "apple_m2_max": 16766.5, "apple_m2_ultra": 33532.9, "apple_m3_max": 16766.5, "apple_m3_ultra": 34329.3, "apple_m4": 5029.9, "apple_m4_max": 22886.2, "apple_m4_pro": 11443.1, "nvidia_rtx_3060_12gb": 15089.8, "nvidia_rtx_3090": 39233.5, "nvidia_rtx_4060_ti_16gb": 12071.9, "nvidia_rtx_4070_ti_super": 28167.7, "nvidia_rtx_4080_super": 30850.3, "nvidia_rtx_4090": 42251.5, "nvidia_rtx_5080": 40239.5, "nvidia_rtx_5090": 75113.8}, "model_id": "sentence-transformers/paraphrase-minilm-l12-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 24.0, "mteb_classification": 57.5, "mteb_clustering": 34.5, "mteb_overall": 50.5, "mteb_pair_classification": 75.5, "mteb_reranking": 44.0, "mteb_retrieval": 41.5, "mteb_sts": 72.0, "mteb_summarization": 23.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "paraphrase MiniLM L3 v2", "fits": {"amd_rx_7900_xt": 65882.4, "amd_rx_7900_xtx": 79058.8, "amd_rx_9070_xt": 53117.6, "apple_m1_max": 32941.2, "apple_m2_max": 32941.2, "apple_m2_ultra": 65882.4, "apple_m3_max": 32941.2, "apple_m3_ultra": 67447.1, "apple_m4": 9882.4, "apple_m4_max": 44964.7, "apple_m4_pro": 22482.4, "nvidia_rtx_3060_12gb": 29647.1, "nvidia_rtx_3090": 77082.4, "nvidia_rtx_4060_ti_16gb": 23717.6, "nvidia_rtx_4070_ti_super": 55341.2, "nvidia_rtx_4080_super": 60611.8, "nvidia_rtx_4090": 83011.8, "nvidia_rtx_5080": 79058.8, "nvidia_rtx_5090": 147576.5}, "model_id": "sentence-transformers/paraphrase-minilm-l3-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 26.5, "mteb_classification": 59.8, "mteb_clustering": 36.8, "mteb_overall": 52.8, "mteb_pair_classification": 77.5, "mteb_reranking": 46.5, "mteb_retrieval": 44.0, "mteb_sts": 74.5, "mteb_summarization": 24.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "paraphrase MiniLM L6 v2", "fits": {"amd_rx_7900_xt": 49339.2, "amd_rx_7900_xtx": 59207.0, "amd_rx_9070_xt": 39779.7, "apple_m1_max": 24669.6, "apple_m2_max": 24669.6, "apple_m2_ultra": 49339.2, "apple_m3_max": 24669.6, "apple_m3_ultra": 50511.0, "apple_m4": 7400.9, "apple_m4_max": 33674.0, "apple_m4_pro": 16837.0, "nvidia_rtx_3060_12gb": 22202.6, "nvidia_rtx_3090": 57726.9, "nvidia_rtx_4060_ti_16gb": 17762.1, "nvidia_rtx_4070_ti_super": 41444.9, "nvidia_rtx_4080_super": 45392.1, "nvidia_rtx_4090": 62167.4, "nvidia_rtx_5080": 59207.0, "nvidia_rtx_5090": 110519.8}, "model_id": "sentence-transformers/paraphrase-minilm-l6-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 30.5, "mteb_classification": 62.2, "mteb_clustering": 39.2, "mteb_overall": 55.5, "mteb_pair_classification": 80.0, "mteb_reranking": 49.0, "mteb_retrieval": 47.0, "mteb_sts": 77.8, "mteb_summarization": 26.2}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "paraphrase mpnet base v2", "fits": {"amd_rx_7900_xt": 10275.2, "amd_rx_7900_xtx": 12330.3, "amd_rx_9070_xt": 8284.4, "apple_m1_max": 5137.6, "apple_m2_max": 5137.6, "apple_m2_ultra": 10275.2, "apple_m3_max": 5137.6, "apple_m3_ultra": 10519.3, "apple_m4": 1541.3, "apple_m4_max": 7012.8, "apple_m4_pro": 3506.4, "nvidia_rtx_3060_12gb": 4623.9, "nvidia_rtx_3090": 12022.0, "nvidia_rtx_4060_ti_16gb": 3699.1, "nvidia_rtx_4070_ti_super": 8631.2, "nvidia_rtx_4080_super": 9453.2, "nvidia_rtx_4090": 12946.8, "nvidia_rtx_5080": 12330.3, "nvidia_rtx_5090": 23016.5}, "model_id": "sentence-transformers/paraphrase-mpnet-base-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 42.0, "mteb_classification": 60.5, "mteb_clustering": 37.5, "mteb_overall": 53.5, "mteb_pair_classification": 78.0, "mteb_reranking": 47.0, "mteb_retrieval": 44.8, "mteb_sts": 75.0, "mteb_summarization": 25.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "paraphrase multilingual MiniLM L12 v2", "fits": {"amd_rx_7900_xt": 9491.5, "amd_rx_7900_xtx": 11389.8, "amd_rx_9070_xt": 7652.5, "apple_m1_max": 4745.8, "apple_m2_max": 4745.8, "apple_m2_ultra": 9491.5, "apple_m3_max": 4745.8, "apple_m3_ultra": 9716.9, "apple_m4": 1423.7, "apple_m4_max": 6478.0, "apple_m4_pro": 3239.0, "nvidia_rtx_3060_12gb": 4271.2, "nvidia_rtx_3090": 11105.1, "nvidia_rtx_4060_ti_16gb": 3416.9, "nvidia_rtx_4070_ti_super": 7972.9, "nvidia_rtx_4080_super": 8732.2, "nvidia_rtx_4090": 11959.3, "nvidia_rtx_5080": 11389.8, "nvidia_rtx_5090": 21261.0}, "model_id": "sentence-transformers/paraphrase-multilingual-minilm-l12-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 43.5, "mteb_classification": 61.5, "mteb_clustering": 38.5, "mteb_overall": 54.8, "mteb_pair_classification": 79.0, "mteb_reranking": 48.2, "mteb_retrieval": 46.2, "mteb_sts": 76.5, "mteb_summarization": 25.8}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "paraphrase multilingual mpnet base v2", "fits": {"amd_rx_7900_xt": 4028.8, "amd_rx_7900_xtx": 4834.5, "amd_rx_9070_xt": 3248.2, "apple_m1_max": 2014.4, "apple_m2_max": 2014.4, "apple_m2_ultra": 4028.8, "apple_m3_max": 2014.4, "apple_m3_ultra": 4124.5, "apple_m4": 604.3, "apple_m4_max": 2749.6, "apple_m4_pro": 1374.8, "nvidia_rtx_3060_12gb": 1812.9, "nvidia_rtx_3090": 4713.7, "nvidia_rtx_4060_ti_16gb": 1450.4, "nvidia_rtx_4070_ti_super": 3384.2, "nvidia_rtx_4080_super": 3706.5, "nvidia_rtx_4090": 5076.3, "nvidia_rtx_5080": 4834.5, "nvidia_rtx_5090": 9024.5}, "model_id": "sentence-transformers/paraphrase-multilingual-mpnet-base-v2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Sentence Transformers", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Skywork o1 Open PRM Qwen 2.5 1.5B", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "skywork/skywork-o1-open-prm-qwen-2-5-1-5b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Skywork o1 Open PRM Qwen 2.5 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "skywork/skywork-o1-open-prm-qwen-2-5-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Skywork R1V 38B", "fits": {"apple_m1_max": 14.7, "apple_m2_max": 14.7, "apple_m2_ultra": 29.5, "apple_m3_max": 14.7, "apple_m3_ultra": 30.2, "apple_m4": 4.4, "apple_m4_max": 20.1, "apple_m4_pro": 10.1, "nvidia_rtx_5090": 66.0}, "model_id": "skywork/skywork-r1v-38b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.1, "display_name": "Skywork Reward Llama 3.1 8B v0.2", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "skywork/skywork-reward-llama-3-1-8b-v0-2", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.1, "display_name": "Skywork Reward V2 Llama 3.1 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "skywork/skywork-reward-v2-llama-3-1-8b", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.04, "display_name": "Skywork Reward V2 Llama 3.2 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "skywork/skywork-reward-v2-llama-3-2-1b", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.03, "display_name": "Skywork Reward V2 Qwen3 0.6B", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "skywork/skywork-reward-v2-qwen3-0-6b", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.04, "display_name": "Skywork Reward V2 Qwen3 1.7B", "fits": {"amd_rx_7900_xt": 658.8, "amd_rx_7900_xtx": 790.6, "amd_rx_9070_xt": 531.2, "apple_m1_max": 329.4, "apple_m2_max": 329.4, "apple_m2_ultra": 658.8, "apple_m3_max": 329.4, "apple_m3_ultra": 674.5, "apple_m4": 98.8, "apple_m4_max": 449.6, "apple_m4_pro": 224.8, "nvidia_rtx_3060_12gb": 296.5, "nvidia_rtx_3090": 770.8, "nvidia_rtx_4060_ti_16gb": 237.2, "nvidia_rtx_4070_ti_super": 553.4, "nvidia_rtx_4080_super": 606.1, "nvidia_rtx_4090": 830.1, "nvidia_rtx_5080": 790.6, "nvidia_rtx_5090": 1475.8}, "model_id": "skywork/skywork-reward-v2-qwen3-1-7b", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {"artificial_analysis_quality_index": 66.0, "artificial_analysis_speed_index": 88.0}, "capability_tiers": {}, "context_window": null, "cost_input": 0.1, "display_name": "Skywork Reward V2 Qwen3 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "skywork/skywork-reward-v2-qwen3-8b", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Skywork VL Reward 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "skywork/skywork-vl-reward-7b", "model_subtypes": [], "model_type": "reward-model", "open_weights": true, "provider": "Skywork", "scores_as_of": null}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Arctic LSTM Speculator Llama 3.1 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "snowflake/arctic-lstm-speculator-llama-3-1-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Snowflake", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Arctic Text2SQL R1 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "snowflake/arctic-text2sql-r1-7b", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Snowflake", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Llama 3.1 SwiftKV 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "snowflake/llama-3-1-swiftkv-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Snowflake", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Llama 3.3 SwiftKV 70B Instruct", "fits": {"apple_m1_max": 8.0, "apple_m2_max": 8.0, "apple_m2_ultra": 16.0, "apple_m3_max": 8.0, "apple_m3_ultra": 16.4, "apple_m4_max": 10.9, "apple_m4_pro": 5.5}, "model_id": "snowflake/llama-3-3-swiftkv-70b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Snowflake", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "snowflake arctic base", "fits": {"apple_m3_ultra": 2.4}, "model_id": "snowflake/snowflake-arctic-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Snowflake", "scores_as_of": null}, {"benchmark_scores": {"beir": 56.0, "miracl": 64.9, "mteb_classification": 71.0, "mteb_clustering": 49.8, "mteb_overall": 65.5, "mteb_pair_classification": 85.5, "mteb_reranking": 58.0, "mteb_retrieval": 63.0, "mteb_sts": 83.5, "mteb_summarization": 30.2}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed l v2.0", "fits": {"amd_rx_7900_xt": 1971.8, "amd_rx_7900_xtx": 2366.2, "amd_rx_9070_xt": 1589.8, "apple_m1_max": 985.9, "apple_m2_max": 985.9, "apple_m2_ultra": 1971.8, "apple_m3_max": 985.9, "apple_m3_ultra": 2018.7, "apple_m4": 295.8, "apple_m4_max": 1345.8, "apple_m4_pro": 672.9, "nvidia_rtx_3060_12gb": 887.3, "nvidia_rtx_3090": 2307.0, "nvidia_rtx_4060_ti_16gb": 709.9, "nvidia_rtx_4070_ti_super": 1656.3, "nvidia_rtx_4080_super": 1814.1, "nvidia_rtx_4090": 2484.5, "nvidia_rtx_5080": 2366.2, "nvidia_rtx_5090": 4416.9}, "model_id": "snowflake/snowflake-arctic-embed-l-v2-0", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 53.8, "miracl": 52.0, "mteb_classification": 69.1, "mteb_clustering": 47.2, "mteb_overall": 63.2, "mteb_pair_classification": 84.5, "mteb_reranking": 56.5, "mteb_retrieval": 60.5, "mteb_sts": 82.0, "mteb_summarization": 29.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed l", "fits": {"amd_rx_7900_xt": 3343.3, "amd_rx_7900_xtx": 4011.9, "amd_rx_9070_xt": 2695.5, "apple_m1_max": 1671.6, "apple_m2_max": 1671.6, "apple_m2_ultra": 3343.3, "apple_m3_max": 1671.6, "apple_m3_ultra": 3422.7, "apple_m4": 501.5, "apple_m4_max": 2281.8, "apple_m4_pro": 1140.9, "nvidia_rtx_3060_12gb": 1504.5, "nvidia_rtx_3090": 3911.6, "nvidia_rtx_4060_ti_16gb": 1203.6, "nvidia_rtx_4070_ti_super": 2808.4, "nvidia_rtx_4080_super": 3075.8, "nvidia_rtx_4090": 4212.5, "nvidia_rtx_5080": 4011.9, "nvidia_rtx_5090": 7489.0}, "model_id": "snowflake/snowflake-arctic-embed-l", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 51.0, "miracl": 48.0, "mteb_classification": 67.0, "mteb_clustering": 44.8, "mteb_overall": 61.0, "mteb_pair_classification": 82.8, "mteb_reranking": 54.2, "mteb_retrieval": 57.5, "mteb_sts": 80.0, "mteb_summarization": 28.2}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed m long", "fits": {"amd_rx_7900_xt": 8175.2, "amd_rx_7900_xtx": 9810.2, "amd_rx_9070_xt": 6591.2, "apple_m1_max": 4087.6, "apple_m2_max": 4087.6, "apple_m2_ultra": 8175.2, "apple_m3_max": 4087.6, "apple_m3_ultra": 8369.3, "apple_m4": 1226.3, "apple_m4_max": 5579.6, "apple_m4_pro": 2789.8, "nvidia_rtx_3060_12gb": 3678.8, "nvidia_rtx_3090": 9565.0, "nvidia_rtx_4060_ti_16gb": 2943.1, "nvidia_rtx_4070_ti_super": 6867.2, "nvidia_rtx_4080_super": 7521.2, "nvidia_rtx_4090": 10300.7, "nvidia_rtx_5080": 9810.2, "nvidia_rtx_5090": 18312.4}, "model_id": "snowflake/snowflake-arctic-embed-m-long", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 51.5, "miracl": 50.5, "mteb_classification": 67.2, "mteb_clustering": 45.2, "mteb_overall": 61.5, "mteb_pair_classification": 83.0, "mteb_reranking": 55.0, "mteb_retrieval": 58.0, "mteb_sts": 80.5, "mteb_summarization": 28.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed m v1.5", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "snowflake/snowflake-arctic-embed-m-v1-5", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 53.0, "miracl": 59.2, "mteb_classification": 68.5, "mteb_clustering": 46.5, "mteb_overall": 62.8, "mteb_pair_classification": 84.0, "mteb_reranking": 56.0, "mteb_retrieval": 59.5, "mteb_sts": 81.5, "mteb_summarization": 29.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed m v2.0", "fits": {"amd_rx_7900_xt": 3672.1, "amd_rx_7900_xtx": 4406.6, "amd_rx_9070_xt": 2960.7, "apple_m1_max": 1836.1, "apple_m2_max": 1836.1, "apple_m2_ultra": 3672.1, "apple_m3_max": 1836.1, "apple_m3_ultra": 3759.3, "apple_m4": 550.8, "apple_m4_max": 2506.2, "apple_m4_pro": 1253.1, "nvidia_rtx_3060_12gb": 1652.5, "nvidia_rtx_3090": 4296.4, "nvidia_rtx_4060_ti_16gb": 1322.0, "nvidia_rtx_4070_ti_super": 3084.6, "nvidia_rtx_4080_super": 3378.4, "nvidia_rtx_4090": 4626.9, "nvidia_rtx_5080": 4406.6, "nvidia_rtx_5090": 8225.6}, "model_id": "snowflake/snowflake-arctic-embed-m-v2-0", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 50.5, "miracl": 48.5, "mteb_classification": 66.5, "mteb_clustering": 44.2, "mteb_overall": 60.5, "mteb_pair_classification": 83.0, "mteb_reranking": 54.5, "mteb_retrieval": 56.8, "mteb_sts": 80.0, "mteb_summarization": 28.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed m", "fits": {"amd_rx_7900_xt": 10181.8, "amd_rx_7900_xtx": 12218.2, "amd_rx_9070_xt": 8209.1, "apple_m1_max": 5090.9, "apple_m2_max": 5090.9, "apple_m2_ultra": 10181.8, "apple_m3_max": 5090.9, "apple_m3_ultra": 10423.6, "apple_m4": 1527.3, "apple_m4_max": 6949.1, "apple_m4_pro": 3474.5, "nvidia_rtx_3060_12gb": 4581.8, "nvidia_rtx_3090": 11912.7, "nvidia_rtx_4060_ti_16gb": 3665.5, "nvidia_rtx_4070_ti_super": 8552.7, "nvidia_rtx_4080_super": 9367.3, "nvidia_rtx_4090": 12829.1, "nvidia_rtx_5080": 12218.2, "nvidia_rtx_5090": 22807.3}, "model_id": "snowflake/snowflake-arctic-embed-m", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 47.2, "miracl": 44.0, "mteb_classification": 64.0, "mteb_clustering": 41.8, "mteb_overall": 57.8, "mteb_pair_classification": 81.5, "mteb_reranking": 52.5, "mteb_retrieval": 53.5, "mteb_sts": 78.0, "mteb_summarization": 27.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed s", "fits": {"amd_rx_7900_xt": 33939.4, "amd_rx_7900_xtx": 40727.3, "amd_rx_9070_xt": 27363.6, "apple_m1_max": 16969.7, "apple_m2_max": 16969.7, "apple_m2_ultra": 33939.4, "apple_m3_max": 16969.7, "apple_m3_ultra": 34745.5, "apple_m4": 5090.9, "apple_m4_max": 23163.6, "apple_m4_pro": 11581.8, "nvidia_rtx_3060_12gb": 15272.7, "nvidia_rtx_3090": 39709.1, "nvidia_rtx_4060_ti_16gb": 12218.2, "nvidia_rtx_4070_ti_super": 28509.1, "nvidia_rtx_4080_super": 31224.2, "nvidia_rtx_4090": 42763.6, "nvidia_rtx_5080": 40727.3, "nvidia_rtx_5090": 76024.2}, "model_id": "snowflake/snowflake-arctic-embed-s", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 44.5, "miracl": 40.0, "mteb_classification": 62.0, "mteb_clustering": 39.5, "mteb_overall": 55.2, "mteb_pair_classification": 79.5, "mteb_reranking": 50.5, "mteb_retrieval": 50.5, "mteb_sts": 76.0, "mteb_summarization": 25.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "snowflake arctic embed xs", "fits": {"amd_rx_7900_xt": 50909.1, "amd_rx_7900_xtx": 61090.9, "amd_rx_9070_xt": 41045.5, "apple_m1_max": 25454.5, "apple_m2_max": 25454.5, "apple_m2_ultra": 50909.1, "apple_m3_max": 25454.5, "apple_m3_ultra": 52118.2, "apple_m4": 7636.4, "apple_m4_max": 34745.5, "apple_m4_pro": 17372.7, "nvidia_rtx_3060_12gb": 22909.1, "nvidia_rtx_3090": 59563.6, "nvidia_rtx_4060_ti_16gb": 18327.3, "nvidia_rtx_4070_ti_super": 42763.6, "nvidia_rtx_4080_super": 46836.4, "nvidia_rtx_4090": 64145.5, "nvidia_rtx_5080": 61090.9, "nvidia_rtx_5090": 114036.4}, "model_id": "snowflake/snowflake-arctic-embed-xs", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Snowflake", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "snowflake arctic instruct", "fits": {"apple_m3_ultra": 2.4}, "model_id": "snowflake/snowflake-arctic-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Snowflake", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.002, "display_name": "sd turbo", "fits": {"amd_rx_7900_xt": 1302.3, "amd_rx_7900_xtx": 1562.8, "amd_rx_9070_xt": 1050.0, "apple_m1_max": 651.2, "apple_m2_max": 651.2, "apple_m2_ultra": 1302.3, "apple_m3_max": 651.2, "apple_m3_ultra": 1333.3, "apple_m4": 195.3, "apple_m4_max": 888.8, "apple_m4_pro": 444.4, "nvidia_rtx_3060_12gb": 586.0, "nvidia_rtx_3090": 1523.7, "nvidia_rtx_4060_ti_16gb": 468.8, "nvidia_rtx_4070_ti_super": 1094.0, "nvidia_rtx_4080_super": 1198.1, "nvidia_rtx_4090": 1640.9, "nvidia_rtx_5080": 1562.8, "nvidia_rtx_5090": 2917.2}, "model_id": "stability/sd-turbo", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "sd vae ft ema", "fits": {"amd_rx_7900_xt": 13333.3, "amd_rx_7900_xtx": 16000.0, "amd_rx_9070_xt": 10750.0, "apple_m1_max": 6666.7, "apple_m2_max": 6666.7, "apple_m2_ultra": 13333.3, "apple_m3_max": 6666.7, "apple_m3_ultra": 13650.0, "apple_m4": 2000.0, "apple_m4_max": 9100.0, "apple_m4_pro": 4550.0, "nvidia_rtx_3060_12gb": 6000.0, "nvidia_rtx_3090": 15600.0, "nvidia_rtx_4060_ti_16gb": 4800.0, "nvidia_rtx_4070_ti_super": 11200.0, "nvidia_rtx_4080_super": 12266.7, "nvidia_rtx_4090": 16800.0, "nvidia_rtx_5080": 16000.0, "nvidia_rtx_5090": 29866.7}, "model_id": "stability/sd-vae-ft-ema", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "sd vae ft mse", "fits": {"amd_rx_7900_xt": 13333.3, "amd_rx_7900_xtx": 16000.0, "amd_rx_9070_xt": 10750.0, "apple_m1_max": 6666.7, "apple_m2_max": 6666.7, "apple_m2_ultra": 13333.3, "apple_m3_max": 6666.7, "apple_m3_ultra": 13650.0, "apple_m4": 2000.0, "apple_m4_max": 9100.0, "apple_m4_pro": 4550.0, "nvidia_rtx_3060_12gb": 6000.0, "nvidia_rtx_3090": 15600.0, "nvidia_rtx_4060_ti_16gb": 4800.0, "nvidia_rtx_4070_ti_super": 11200.0, "nvidia_rtx_4080_super": 12266.7, "nvidia_rtx_4090": 16800.0, "nvidia_rtx_5080": 16000.0, "nvidia_rtx_5090": 29866.7}, "model_id": "stability/sd-vae-ft-mse", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "sd x2 latent upscaler", "fits": {"amd_rx_7900_xt": 2800.0, "amd_rx_7900_xtx": 3360.0, "amd_rx_9070_xt": 2257.5, "apple_m1_max": 1400.0, "apple_m2_max": 1400.0, "apple_m2_ultra": 2800.0, "apple_m3_max": 1400.0, "apple_m3_ultra": 2866.5, "apple_m4": 420.0, "apple_m4_max": 1911.0, "apple_m4_pro": 955.5, "nvidia_rtx_3060_12gb": 1260.0, "nvidia_rtx_3090": 3276.0, "nvidia_rtx_4060_ti_16gb": 1008.0, "nvidia_rtx_4070_ti_super": 2352.0, "nvidia_rtx_4080_super": 2576.0, "nvidia_rtx_4090": 3528.0, "nvidia_rtx_5080": 3360.0, "nvidia_rtx_5090": 6272.0}, "model_id": "stability/sd-x2-latent-upscaler", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.002, "display_name": "sdxl turbo ryzen ai", "fits": {"amd_rx_7900_xt": 320.0, "amd_rx_7900_xtx": 384.0, "amd_rx_9070_xt": 258.0, "apple_m1_max": 160.0, "apple_m2_max": 160.0, "apple_m2_ultra": 320.0, "apple_m3_max": 160.0, "apple_m3_ultra": 327.6, "apple_m4": 48.0, "apple_m4_max": 218.4, "apple_m4_pro": 109.2, "nvidia_rtx_3060_12gb": 144.0, "nvidia_rtx_3090": 374.4, "nvidia_rtx_4060_ti_16gb": 115.2, "nvidia_rtx_4070_ti_super": 268.8, "nvidia_rtx_4080_super": 294.4, "nvidia_rtx_4090": 403.2, "nvidia_rtx_5080": 384.0, "nvidia_rtx_5090": 716.8}, "model_id": "stability/sdxl-turbo-ryzen-ai", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.002, "display_name": "sdxl turbo", "fits": {"amd_rx_7900_xt": 320.0, "amd_rx_7900_xtx": 384.0, "amd_rx_9070_xt": 258.0, "apple_m1_max": 160.0, "apple_m2_max": 160.0, "apple_m2_ultra": 320.0, "apple_m3_max": 160.0, "apple_m3_ultra": 327.6, "apple_m4": 48.0, "apple_m4_max": 218.4, "apple_m4_pro": 109.2, "nvidia_rtx_3060_12gb": 144.0, "nvidia_rtx_3090": 374.4, "nvidia_rtx_4060_ti_16gb": 115.2, "nvidia_rtx_4070_ti_super": 268.8, "nvidia_rtx_4080_super": 294.4, "nvidia_rtx_4090": 403.2, "nvidia_rtx_5080": 384.0, "nvidia_rtx_5090": 716.8}, "model_id": "stability/sdxl-turbo", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": 0.002, "display_name": "sdxl vae", "fits": {"amd_rx_7900_xt": 13333.3, "amd_rx_7900_xtx": 16000.0, "amd_rx_9070_xt": 10750.0, "apple_m1_max": 6666.7, "apple_m2_max": 6666.7, "apple_m2_ultra": 13333.3, "apple_m3_max": 6666.7, "apple_m3_ultra": 13650.0, "apple_m4": 2000.0, "apple_m4_max": 9100.0, "apple_m4_pro": 4550.0, "nvidia_rtx_3060_12gb": 6000.0, "nvidia_rtx_3090": 15600.0, "nvidia_rtx_4060_ti_16gb": 4800.0, "nvidia_rtx_4070_ti_super": 11200.0, "nvidia_rtx_4080_super": 12266.7, "nvidia_rtx_4090": 16800.0, "nvidia_rtx_5080": 16000.0, "nvidia_rtx_5090": 29866.7}, "model_id": "stability/sdxl-vae", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "stable audio open 1.0", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "stability/stable-audio-open-1-0", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "stable cascade", "fits": {"amd_rx_7900_xt": 219.6, "amd_rx_7900_xtx": 263.5, "amd_rx_9070_xt": 177.1, "apple_m1_max": 109.8, "apple_m2_max": 109.8, "apple_m2_ultra": 219.6, "apple_m3_max": 109.8, "apple_m3_ultra": 224.8, "apple_m4": 32.9, "apple_m4_max": 149.9, "apple_m4_pro": 74.9, "nvidia_rtx_3060_12gb": 98.8, "nvidia_rtx_3090": 256.9, "nvidia_rtx_4060_ti_16gb": 79.1, "nvidia_rtx_4070_ti_super": 184.5, "nvidia_rtx_4080_super": 202.0, "nvidia_rtx_4090": 276.7, "nvidia_rtx_5080": 263.5, "nvidia_rtx_5090": 491.9}, "model_id": "stability/stable-cascade", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 16384, "cost_input": null, "display_name": "stable code 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "stability/stable-code-3b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "stable diffusion 3.5 large controlnet canny", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "stability/stable-diffusion-3-5-large-controlnet-canny", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "stable diffusion 3.5 large controlnet depth", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "stability/stable-diffusion-3-5-large-controlnet-depth", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "stable diffusion 3.5 large turbo", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "stability/stable-diffusion-3-5-large-turbo", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "stable diffusion 3.5 large", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "stability/stable-diffusion-3-5-large", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "stable diffusion 3.5 medium", "fits": {"amd_rx_7900_xt": 448.0, "amd_rx_7900_xtx": 537.6, "amd_rx_9070_xt": 361.2, "apple_m1_max": 224.0, "apple_m2_max": 224.0, "apple_m2_ultra": 448.0, "apple_m3_max": 224.0, "apple_m3_ultra": 458.6, "apple_m4": 67.2, "apple_m4_max": 305.8, "apple_m4_pro": 152.9, "nvidia_rtx_3060_12gb": 201.6, "nvidia_rtx_3090": 524.2, "nvidia_rtx_4060_ti_16gb": 161.3, "nvidia_rtx_4070_ti_super": 376.3, "nvidia_rtx_4080_super": 412.2, "nvidia_rtx_4090": 564.5, "nvidia_rtx_5080": 537.6, "nvidia_rtx_5090": 1003.5}, "model_id": "stability/stable-diffusion-3-5-medium", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": 0.035, "display_name": "stable diffusion 3 medium diffusers", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "stability/stable-diffusion-3-medium-diffusers", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "stable diffusion x4 upscaler", "fits": {"amd_rx_7900_xt": 1302.3, "amd_rx_7900_xtx": 1562.8, "amd_rx_9070_xt": 1050.0, "apple_m1_max": 651.2, "apple_m2_max": 651.2, "apple_m2_ultra": 1302.3, "apple_m3_max": 651.2, "apple_m3_ultra": 1333.3, "apple_m4": 195.3, "apple_m4_max": 888.8, "apple_m4_pro": 444.4, "nvidia_rtx_3060_12gb": 586.0, "nvidia_rtx_3090": 1523.7, "nvidia_rtx_4060_ti_16gb": 468.8, "nvidia_rtx_4070_ti_super": 1094.0, "nvidia_rtx_4080_super": 1198.1, "nvidia_rtx_4090": 1640.9, "nvidia_rtx_5080": 1562.8, "nvidia_rtx_5090": 2917.2}, "model_id": "stability/stable-diffusion-x4-upscaler", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "stable diffusion xl base 1.0", "fits": {"amd_rx_7900_xt": 320.0, "amd_rx_7900_xtx": 384.0, "amd_rx_9070_xt": 258.0, "apple_m1_max": 160.0, "apple_m2_max": 160.0, "apple_m2_ultra": 320.0, "apple_m3_max": 160.0, "apple_m3_ultra": 327.6, "apple_m4": 48.0, "apple_m4_max": 218.4, "apple_m4_pro": 109.2, "nvidia_rtx_3060_12gb": 144.0, "nvidia_rtx_3090": 374.4, "nvidia_rtx_4060_ti_16gb": 115.2, "nvidia_rtx_4070_ti_super": 268.8, "nvidia_rtx_4080_super": 294.4, "nvidia_rtx_4090": 403.2, "nvidia_rtx_5080": 384.0, "nvidia_rtx_5090": 716.8}, "model_id": "stability/stable-diffusion-xl-base-1-0", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "stable diffusion xl refiner 1.0", "fits": {"amd_rx_7900_xt": 320.0, "amd_rx_7900_xtx": 384.0, "amd_rx_9070_xt": 258.0, "apple_m1_max": 160.0, "apple_m2_max": 160.0, "apple_m2_ultra": 320.0, "apple_m3_max": 160.0, "apple_m3_ultra": 327.6, "apple_m4": 48.0, "apple_m4_max": 218.4, "apple_m4_pro": 109.2, "nvidia_rtx_3060_12gb": 144.0, "nvidia_rtx_3090": 374.4, "nvidia_rtx_4060_ti_16gb": 115.2, "nvidia_rtx_4070_ti_super": 268.8, "nvidia_rtx_4080_super": 294.4, "nvidia_rtx_4090": 403.2, "nvidia_rtx_5080": 384.0, "nvidia_rtx_5090": 716.8}, "model_id": "stability/stable-diffusion-xl-refiner-1-0", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "stable fast 3d", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "stability/stable-fast-3d", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "stable video diffusion img2vid xt 1 1", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "stability/stable-video-diffusion-img2vid-xt-1-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "stable video diffusion img2vid xt", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "stability/stable-video-diffusion-img2vid-xt", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "stable video diffusion img2vid", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "stability/stable-video-diffusion-img2vid", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "stable virtual camera", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "stability/stable-virtual-camera", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 33.8, "gpqa_diamond": 24.8, "ifeval": 11.6, "math_500": 0.8, "mmlu_pro": 14.6, "musr": 38.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "stablelm 2 1 6B", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "stability/stablelm-2-1-6b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Stability AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 35.0, "gpqa_diamond": 23.7, "ifeval": 22.0, "math_500": 1.1, "mmlu_pro": 16.7, "musr": 37.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "stablelm 3B 4e1t", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "stability/stablelm-3b-4e1t", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Stability AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "stablelm base alpha 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "stability/stablelm-base-alpha-3b", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "stablelm base alpha 7B v2", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "stability/stablelm-base-alpha-7b-v2", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 38.7, "gpqa_diamond": 23.9, "ifeval": 36.8, "math_500": 4.3, "mmlu_pro": 17.7, "musr": 41.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "stablelm zephyr 3B", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "stability/stablelm-zephyr-3b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "TripoSR", "fits": {"amd_rx_7900_xt": 2800.0, "amd_rx_7900_xtx": 3360.0, "amd_rx_9070_xt": 2257.5, "apple_m1_max": 1400.0, "apple_m2_max": 1400.0, "apple_m2_ultra": 2800.0, "apple_m3_max": 1400.0, "apple_m3_ultra": 2866.5, "apple_m4": 420.0, "apple_m4_max": 1911.0, "apple_m4_pro": 955.5, "nvidia_rtx_3060_12gb": 1260.0, "nvidia_rtx_3090": 3276.0, "nvidia_rtx_4060_ti_16gb": 1008.0, "nvidia_rtx_4070_ti_super": 2352.0, "nvidia_rtx_4080_super": 2576.0, "nvidia_rtx_4090": 3528.0, "nvidia_rtx_5080": 3360.0, "nvidia_rtx_5090": 6272.0}, "model_id": "stability/triposr", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Stability AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 2.05, "display_name": "Step 1 (32K)", "fits": {}, "model_id": "stepfun/step-1-32k", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "StepFun", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 16384, "cost_input": 5.21, "display_name": "Step 2 (16K)", "fits": {}, "model_id": "stepfun/step-2-16k", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "StepFun", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.096, "display_name": "Step 3.5 Flash", "fits": {}, "model_id": "stepfun/step-3-5-flash", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "StepFun", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Llama 3.1 AlternateTokenizer", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "teknium/llama-3-1-alternatetokenizer", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Teknium", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Mistral Trismegistus 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "teknium/mistral-trismegistus-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Teknium", "scores_as_of": null}, {"benchmark_scores": {"bbh": 42.1, "gpqa_diamond": 27.3, "ifeval": 26.7, "math_500": 1.2, "mmlu_pro": 23.9, "musr": 40.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "OpenHermes 13B", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "teknium/openhermes-13b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Teknium", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 48.7, "gpqa_diamond": 28.4, "ifeval": 55.7, "math_500": 5.1, "mmlu_pro": 30.5, "musr": 42.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "OpenHermes 2.5 Mistral 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "teknium/openhermes-2-5-mistral-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Teknium", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 49.5, "gpqa_diamond": 28.4, "ifeval": 52.9, "math_500": 4.5, "mmlu_pro": 29.3, "musr": 45.2}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "OpenHermes 2 Mistral 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "teknium/openhermes-2-mistral-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Teknium", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "DepthCrafter", "fits": {}, "model_id": "tencent/depthcrafter", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan 0.5B Pretrain", "fits": {"amd_rx_7900_xt": 2240.0, "amd_rx_7900_xtx": 2688.0, "amd_rx_9070_xt": 1806.0, "apple_m1_max": 1120.0, "apple_m2_max": 1120.0, "apple_m2_ultra": 2240.0, "apple_m3_max": 1120.0, "apple_m3_ultra": 2293.2, "apple_m4": 336.0, "apple_m4_max": 1528.8, "apple_m4_pro": 764.4, "nvidia_rtx_3060_12gb": 1008.0, "nvidia_rtx_3090": 2620.8, "nvidia_rtx_4060_ti_16gb": 806.4, "nvidia_rtx_4070_ti_super": 1881.6, "nvidia_rtx_4080_super": 2060.8, "nvidia_rtx_4090": 2822.4, "nvidia_rtx_5080": 2688.0, "nvidia_rtx_5090": 5017.6}, "model_id": "tencent/hunyuan-0-5b-pretrain", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan 1.8B Pretrain", "fits": {"amd_rx_7900_xt": 622.2, "amd_rx_7900_xtx": 746.7, "amd_rx_9070_xt": 501.7, "apple_m1_max": 311.1, "apple_m2_max": 311.1, "apple_m2_ultra": 622.2, "apple_m3_max": 311.1, "apple_m3_ultra": 637.0, "apple_m4": 93.3, "apple_m4_max": 424.7, "apple_m4_pro": 212.3, "nvidia_rtx_3060_12gb": 280.0, "nvidia_rtx_3090": 728.0, "nvidia_rtx_4060_ti_16gb": 224.0, "nvidia_rtx_4070_ti_super": 522.7, "nvidia_rtx_4080_super": 572.4, "nvidia_rtx_4090": 784.0, "nvidia_rtx_5080": 746.7, "nvidia_rtx_5090": 1393.8}, "model_id": "tencent/hunyuan-1-8b-pretrain", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan 4B Pretrain", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "tencent/hunyuan-4b-pretrain", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tencent/hunyuan-7b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan 7B Pretrain", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tencent/hunyuan-7b-pretrain", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan A13B Instruct", "fits": {"amd_rx_7900_xt": 86.2, "amd_rx_7900_xtx": 103.4, "amd_rx_9070_xt": 69.5, "apple_m1_max": 43.1, "apple_m2_max": 43.1, "apple_m2_ultra": 86.2, "apple_m3_max": 43.1, "apple_m3_ultra": 88.2, "apple_m4": 12.9, "apple_m4_max": 58.8, "apple_m4_pro": 29.4, "nvidia_rtx_3060_12gb": 38.8, "nvidia_rtx_3090": 100.8, "nvidia_rtx_4060_ti_16gb": 31.0, "nvidia_rtx_4070_ti_super": 72.4, "nvidia_rtx_4080_super": 79.3, "nvidia_rtx_4090": 108.6, "nvidia_rtx_5080": 103.4, "nvidia_rtx_5090": 193.0}, "model_id": "tencent/hunyuan-a13b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan MT 7B fp8", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tencent/hunyuan-mt-7b-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Hunyuan MT 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tencent/hunyuan-mt-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "Hunyuan3D 2.1", "fits": {}, "model_id": "tencent/hunyuan3d-2-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "Hunyuan3D 2", "fits": {}, "model_id": "tencent/hunyuan3d-2", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "Hunyuan3D 2mini", "fits": {}, "model_id": "tencent/hunyuan3d-2mini", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "Hunyuan3D 2mv", "fits": {}, "model_id": "tencent/hunyuan3d-2mv", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "HunyuanImage 3.0 Instruct", "fits": {}, "model_id": "tencent/hunyuanimage-3-0-instruct", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "HunyuanImage 3.0", "fits": {}, "model_id": "tencent/hunyuanimage-3-0", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "HunyuanOCR", "fits": {}, "model_id": "tencent/hunyuanocr", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "HunyuanWorld 1", "fits": {}, "model_id": "tencent/hunyuanworld-1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "HunyuanWorld Mirror", "fits": {}, "model_id": "tencent/hunyuanworld-mirror", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "HY MT1.5 1.8B GGUF", "fits": {"amd_rx_7900_xt": 622.2, "amd_rx_7900_xtx": 746.7, "amd_rx_9070_xt": 501.7, "apple_m1_max": 311.1, "apple_m2_max": 311.1, "apple_m2_ultra": 622.2, "apple_m3_max": 311.1, "apple_m3_ultra": 637.0, "apple_m4": 93.3, "apple_m4_max": 424.7, "apple_m4_pro": 212.3, "nvidia_rtx_3060_12gb": 280.0, "nvidia_rtx_3090": 728.0, "nvidia_rtx_4060_ti_16gb": 224.0, "nvidia_rtx_4070_ti_super": 522.7, "nvidia_rtx_4080_super": 572.4, "nvidia_rtx_4090": 784.0, "nvidia_rtx_5080": 746.7, "nvidia_rtx_5090": 1393.8}, "model_id": "tencent/hy-mt1-5-1-8b-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "HY MT1.5 1.8B", "fits": {"amd_rx_7900_xt": 622.2, "amd_rx_7900_xtx": 746.7, "amd_rx_9070_xt": 501.7, "apple_m1_max": 311.1, "apple_m2_max": 311.1, "apple_m2_ultra": 622.2, "apple_m3_max": 311.1, "apple_m3_ultra": 637.0, "apple_m4": 93.3, "apple_m4_max": 424.7, "apple_m4_pro": 212.3, "nvidia_rtx_3060_12gb": 280.0, "nvidia_rtx_3090": 728.0, "nvidia_rtx_4060_ti_16gb": 224.0, "nvidia_rtx_4070_ti_super": 522.7, "nvidia_rtx_4080_super": 572.4, "nvidia_rtx_4090": 784.0, "nvidia_rtx_5080": 746.7, "nvidia_rtx_5090": 1393.8}, "model_id": "tencent/hy-mt1-5-1-8b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "HY MT1.5 7B GGUF", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tencent/hy-mt1-5-7b-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "HY MT1.5 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tencent/hy-mt1-5-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "KaLM Embedding Gemma3 12B 2511", "fits": {"amd_rx_7900_xt": 93.3, "amd_rx_7900_xtx": 112.0, "amd_rx_9070_xt": 75.2, "apple_m1_max": 46.7, "apple_m2_max": 46.7, "apple_m2_ultra": 93.3, "apple_m3_max": 46.7, "apple_m3_ultra": 95.5, "apple_m4": 14.0, "apple_m4_max": 63.7, "apple_m4_pro": 31.8, "nvidia_rtx_3060_12gb": 42.0, "nvidia_rtx_3090": 109.2, "nvidia_rtx_4060_ti_16gb": 33.6, "nvidia_rtx_4070_ti_super": 78.4, "nvidia_rtx_4080_super": 85.9, "nvidia_rtx_4090": 117.6, "nvidia_rtx_5080": 112.0, "nvidia_rtx_5090": 209.1}, "model_id": "tencent/kalm-embedding-gemma3-12b-2511", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Penguin Encoder", "fits": {}, "model_id": "tencent/penguin-encoder", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Penguin VL 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "tencent/penguin-vl-2b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "Penguin VL 8B", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "tencent/penguin-vl-8b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "SongGeneration", "fits": {}, "model_id": "tencent/songgeneration", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "WeDLM 8B Base", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "tencent/wedlm-8b-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "WeDLM 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "tencent/wedlm-8b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 4096, "cost_input": null, "display_name": "Youtu LLM 2B Base", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "tencent/youtu-llm-2b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "Tencent", "scores_as_of": null}, {"benchmark_scores": {"bbh": 43.9, "gpqa_diamond": 27.1, "ifeval": 32.6, "math_500": 2.8, "mmlu_pro": 23.9, "musr": 39.9, "winogrande": 78.3}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "falcon 11B", "fits": {"amd_rx_7900_xt": 101.8, "amd_rx_7900_xtx": 122.2, "amd_rx_9070_xt": 82.1, "apple_m1_max": 50.9, "apple_m2_max": 50.9, "apple_m2_ultra": 101.8, "apple_m3_max": 50.9, "apple_m3_ultra": 104.2, "apple_m4": 15.3, "apple_m4_max": 69.5, "apple_m4_pro": 34.7, "nvidia_rtx_3060_12gb": 45.8, "nvidia_rtx_3090": 119.1, "nvidia_rtx_4060_ti_16gb": 36.7, "nvidia_rtx_4070_ti_super": 85.5, "nvidia_rtx_4080_super": 93.7, "nvidia_rtx_4090": 128.3, "nvidia_rtx_5080": 122.2, "nvidia_rtx_5090": 228.1}, "model_id": "tii/falcon-11b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "TII", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 40.5, "gpqa_diamond": 25.0, "gsm8k": 34.3, "ifeval": 24.5, "math_500": 2.0, "mmlu_pro": 22.6, "musr": 37.6}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": 0.6, "display_name": "falcon 40B instruct", "fits": {"apple_m1_max": 14.0, "apple_m2_max": 14.0, "apple_m2_ultra": 28.0, "apple_m3_max": 14.0, "apple_m3_ultra": 28.7, "apple_m4": 4.2, "apple_m4_max": 19.1, "apple_m4_pro": 9.6, "nvidia_rtx_5090": 62.7}, "model_id": "tii/falcon-40b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "TII", "scores_as_of": "2024-07"}, {"benchmark_scores": {"arc_challenge": 61.9, "bbh": 40.2, "gpqa_diamond": 27.3, "hellaswag": 85.3, "ifeval": 25.0, "math_500": 1.8, "mmlu_abstract_algebra": 34.0, "mmlu_anatomy": 53.3, "mmlu_astronomy": 57.2, "mmlu_business_ethics": 54.0, "mmlu_clinical_knowledge": 60.8, "mmlu_college_biology": 66.7, "mmlu_college_chemistry": 45.0, "mmlu_college_computer_science": 50.0, "mmlu_college_mathematics": 39.0, "mmlu_college_medicine": 52.0, "mmlu_college_physics": 27.5, "mmlu_computer_security": 63.0, "mmlu_conceptual_physics": 41.3, "mmlu_econometrics": 33.3, "mmlu_electrical_engineering": 53.1, "mmlu_elementary_mathematics": 33.9, "mmlu_formal_logic": 30.2, "mmlu_global_facts": 32.0, "mmlu_high_school_biology": 67.1, "mmlu_high_school_chemistry": 44.3, "mmlu_high_school_computer_science": 62.0, "mmlu_high_school_european_history": 69.7, "mmlu_high_school_geography": 74.2, "mmlu_high_school_government_and_politics": 77.2, "mmlu_high_school_macroeconomics": 55.4, "mmlu_high_school_mathematics": 31.5, "mmlu_high_school_microeconomics": 55.5, "mmlu_high_school_physics": 31.8, "mmlu_high_school_psychology": 77.6, "mmlu_high_school_statistics": 47.7, "mmlu_high_school_us_history": 71.6, "mmlu_high_school_world_history": 71.3, "mmlu_human_aging": 73.5, "mmlu_human_sexuality": 72.5, "mmlu_international_law": 65.3, "mmlu_jurisprudence": 69.4, "mmlu_logical_fallacies": 65.6, "mmlu_machine_learning": 29.5, "mmlu_management": 78.6, "mmlu_marketing": 79.1, "mmlu_medical_genetics": 66.0, "mmlu_miscellaneous": 75.9, "mmlu_moral_disputes": 62.7, "mmlu_moral_scenarios": 27.0, "mmlu_nutrition": 67.3, "mmlu_philosophy": 66.9, "mmlu_prehistory": 63.3, "mmlu_pro": 25.0, "mmlu_professional_accounting": 43.6, "mmlu_professional_law": 43.2, "mmlu_professional_medicine": 61.0, "mmlu_professional_psychology": 56.5, "mmlu_public_relations": 62.7, "mmlu_security_studies": 66.1, "mmlu_sociology": 79.6, "mmlu_us_foreign_policy": 81.0, "mmlu_virology": 47.6, "mmlu_world_religions": 80.7, "musr": 36.3, "truthfulqa": 41.7}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": 0.6, "display_name": "falcon 40B", "fits": {"apple_m1_max": 14.0, "apple_m2_max": 14.0, "apple_m2_ultra": 28.0, "apple_m3_max": 14.0, "apple_m3_ultra": 28.7, "apple_m4": 4.2, "apple_m4_max": 19.1, "apple_m4_pro": 9.6, "nvidia_rtx_5090": 62.7}, "model_id": "tii/falcon-40b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "TII", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 32.0, "gpqa_diamond": 24.7, "gsm8k": 4.7, "ifeval": 19.7, "math_500": 1.2, "mmlu_pro": 11.6, "musr": 36.3}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": 0.1, "display_name": "falcon 7B instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon-7b-instruct", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "TII", "scores_as_of": "2024-07"}, {"benchmark_scores": {"bbh": 32.9, "gpqa_diamond": 24.5, "gsm8k": 4.6, "ifeval": 18.2, "math_500": 1.0, "mmlu_pro": 11.3, "musr": 37.8}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": 0.1, "display_name": "falcon 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon-7b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "TII", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 0.5B Base", "fits": {"amd_rx_7900_xt": 2240.0, "amd_rx_7900_xtx": 2688.0, "amd_rx_9070_xt": 1806.0, "apple_m1_max": 1120.0, "apple_m2_max": 1120.0, "apple_m2_ultra": 2240.0, "apple_m3_max": 1120.0, "apple_m3_ultra": 2293.2, "apple_m4": 336.0, "apple_m4_max": 1528.8, "apple_m4_pro": 764.4, "nvidia_rtx_3060_12gb": 1008.0, "nvidia_rtx_3090": 2620.8, "nvidia_rtx_4060_ti_16gb": 806.4, "nvidia_rtx_4070_ti_super": 1881.6, "nvidia_rtx_4080_super": 2060.8, "nvidia_rtx_4090": 2822.4, "nvidia_rtx_5080": 2688.0, "nvidia_rtx_5090": 5017.6}, "model_id": "tii/falcon-h1-0-5b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 1.5B Base", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "tii/falcon-h1-1-5b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 1.5B Deep Base", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "tii/falcon-h1-1-5b-deep-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 1.5B Deep Instruct", "fits": {"amd_rx_7900_xt": 746.7, "amd_rx_7900_xtx": 896.0, "amd_rx_9070_xt": 602.0, "apple_m1_max": 373.3, "apple_m2_max": 373.3, "apple_m2_ultra": 746.7, "apple_m3_max": 373.3, "apple_m3_ultra": 764.4, "apple_m4": 112.0, "apple_m4_max": 509.6, "apple_m4_pro": 254.8, "nvidia_rtx_3060_12gb": 336.0, "nvidia_rtx_3090": 873.6, "nvidia_rtx_4060_ti_16gb": 268.8, "nvidia_rtx_4070_ti_super": 627.2, "nvidia_rtx_4080_super": 686.9, "nvidia_rtx_4090": 940.8, "nvidia_rtx_5080": 896.0, "nvidia_rtx_5090": 1672.5}, "model_id": "tii/falcon-h1-1-5b-deep-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 34B Base", "fits": {"amd_rx_7900_xtx": 39.5, "apple_m1_max": 16.5, "apple_m2_max": 16.5, "apple_m2_ultra": 32.9, "apple_m3_max": 16.5, "apple_m3_ultra": 33.7, "apple_m4": 4.9, "apple_m4_max": 22.5, "apple_m4_pro": 11.2, "nvidia_rtx_3090": 38.5, "nvidia_rtx_4090": 41.5, "nvidia_rtx_5090": 73.8}, "model_id": "tii/falcon-h1-34b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 3B Base", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "tii/falcon-h1-3b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 7B Base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon-h1-7b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1 Tiny 90M Instruct", "fits": {"amd_rx_7900_xt": 12444.4, "amd_rx_7900_xtx": 14933.3, "amd_rx_9070_xt": 10033.3, "apple_m1_max": 6222.2, "apple_m2_max": 6222.2, "apple_m2_ultra": 12444.4, "apple_m3_max": 6222.2, "apple_m3_ultra": 12740.0, "apple_m4": 1866.7, "apple_m4_max": 8493.3, "apple_m4_pro": 4246.7, "nvidia_rtx_3060_12gb": 5600.0, "nvidia_rtx_3090": 14560.0, "nvidia_rtx_4060_ti_16gb": 4480.0, "nvidia_rtx_4070_ti_super": 10453.3, "nvidia_rtx_4080_super": 11448.9, "nvidia_rtx_4090": 15680.0, "nvidia_rtx_5080": 14933.3, "nvidia_rtx_5090": 27875.6}, "model_id": "tii/falcon-h1-tiny-90m-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "Falcon H1R 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon-h1r-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {"bbh": 42.8, "gpqa_diamond": 31.0, "ifeval": 33.4, "math_500": 4.5, "mmlu_pro": 23.0, "musr": 42.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "falcon mamba 7B instruct Q4 K M GGUF", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon-mamba-7b-instruct-q4-k-m-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 42.8, "gpqa_diamond": 31.0, "ifeval": 33.4, "math_500": 4.5, "mmlu_pro": 23.0, "musr": 42.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "falcon mamba 7B instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon-mamba-7b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 42.8, "gpqa_diamond": 31.0, "ifeval": 33.4, "math_500": 4.5, "mmlu_pro": 23.0, "musr": 42.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "falcon mamba 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon-mamba-7b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "falcon mamba tiny dev", "fits": {"amd_rx_7900_xt": 8615.4, "amd_rx_7900_xtx": 10338.5, "amd_rx_9070_xt": 6946.2, "apple_m1_max": 4307.7, "apple_m2_max": 4307.7, "apple_m2_ultra": 8615.4, "apple_m3_max": 4307.7, "apple_m3_ultra": 8820.0, "apple_m4": 1292.3, "apple_m4_max": 5880.0, "apple_m4_pro": 2940.0, "nvidia_rtx_3060_12gb": 3876.9, "nvidia_rtx_3090": 10080.0, "nvidia_rtx_4060_ti_16gb": 3101.5, "nvidia_rtx_4070_ti_super": 7236.9, "nvidia_rtx_4080_super": 7926.2, "nvidia_rtx_4090": 10855.4, "nvidia_rtx_5080": 10338.5, "nvidia_rtx_5090": 19298.5}, "model_id": "tii/falcon-mamba-tiny-dev", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 4096, "cost_input": null, "display_name": "Falcon Perception", "fits": {}, "model_id": "tii/falcon-perception", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "falcon rw 1B", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "tii/falcon-rw-1b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {"bbh": 36.5, "gpqa_diamond": 27.8, "ifeval": 69.1, "math_500": 41.2, "mmlu_pro": 40.2, "musr": 11.5}, "capability_tiers": {}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 10B Base", "fits": {"amd_rx_7900_xt": 112.0, "amd_rx_7900_xtx": 134.4, "amd_rx_9070_xt": 90.3, "apple_m1_max": 56.0, "apple_m2_max": 56.0, "apple_m2_ultra": 112.0, "apple_m3_max": 56.0, "apple_m3_ultra": 114.7, "apple_m4": 16.8, "apple_m4_max": 76.4, "apple_m4_pro": 38.2, "nvidia_rtx_3060_12gb": 50.4, "nvidia_rtx_3090": 131.0, "nvidia_rtx_4060_ti_16gb": 40.3, "nvidia_rtx_4070_ti_super": 94.1, "nvidia_rtx_4080_super": 103.0, "nvidia_rtx_4090": 141.1, "nvidia_rtx_5080": 134.4, "nvidia_rtx_5090": 250.9}, "model_id": "tii/falcon3-10b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 36.5, "gpqa_diamond": 27.8, "ifeval": 69.1, "math_500": 41.2, "mmlu_pro": 40.2, "musr": 11.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 10B Instruct", "fits": {"amd_rx_7900_xt": 112.0, "amd_rx_7900_xtx": 134.4, "amd_rx_9070_xt": 90.3, "apple_m1_max": 56.0, "apple_m2_max": 56.0, "apple_m2_ultra": 112.0, "apple_m3_max": 56.0, "apple_m3_ultra": 114.7, "apple_m4": 16.8, "apple_m4_max": 76.4, "apple_m4_pro": 38.2, "nvidia_rtx_3060_12gb": 50.4, "nvidia_rtx_3090": 131.0, "nvidia_rtx_4060_ti_16gb": 40.3, "nvidia_rtx_4070_ti_super": 94.1, "nvidia_rtx_4080_super": 103.0, "nvidia_rtx_4090": 141.1, "nvidia_rtx_5080": 134.4, "nvidia_rtx_5090": 250.9}, "model_id": "tii/falcon3-10b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2026-04"}, {"benchmark_scores": {"bbh": 35.7, "gpqa_diamond": 27.9, "ifeval": 24.3, "math_500": 3.3, "mmlu_pro": 16.1, "musr": 41.5}, "capability_tiers": {}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 1B Base", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "tii/falcon3-1b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 37.4, "gpqa_diamond": 26.7, "ifeval": 55.6, "math_500": 6.3, "mmlu_pro": 18.4, "musr": 41.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 1B Instruct", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "tii/falcon3-1b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 44.2, "gpqa_diamond": 29.7, "ifeval": 27.6, "math_500": 11.8, "mmlu_pro": 28.8, "musr": 37.5}, "capability_tiers": {}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 3B Base", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "tii/falcon3-3b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 47.5, "gpqa_diamond": 28.9, "ifeval": 69.8, "math_500": 25.0, "mmlu_pro": 30.1, "musr": 41.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 3B Instruct", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "tii/falcon3-3b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 51.0, "gpqa_diamond": 34.6, "ifeval": 34.2, "math_500": 19.4, "mmlu_pro": 39.1, "musr": 47.0}, "capability_tiers": {}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 7B Base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon3-7b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 56.3, "gpqa_diamond": 31.0, "ifeval": 76.1, "math_500": 40.9, "mmlu_pro": 40.9, "musr": 48.3}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon3-7b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 46.8, "gpqa_diamond": 30.4, "ifeval": 71.7, "math_500": 30.1, "mmlu_pro": 33.7, "musr": 38.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Falcon3 Mamba 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "tii/falcon3-mamba-7b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "TII", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "siglino moe 0.3 0.6B", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "tii/siglino-moe-0-3-0-6b", "model_subtypes": [], "model_type": "embedding-text", "open_weights": true, "provider": "TII", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 262144, "cost_input": null, "display_name": "Aurora Spec Qwen3 Coder Next FP8", "fits": {"apple_m3_ultra": 2.4}, "model_id": "together/aurora-spec-qwen3-coder-next-fp8", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Together AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "evo 1 131K base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/evo-1-131k-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Together AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "evo 1 8K base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/evo-1-8k-base", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Together AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "GPT JT 6B v0", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "together/gpt-jt-6b-v0", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 33.0, "gpqa_diamond": 26.1, "ifeval": 20.6, "math_500": 1.1, "mmlu_pro": 16.3, "musr": 37.4}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "GPT JT 6B v1", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "together/gpt-jt-6b-v1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "GPT JT Moderation 6B", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "together/gpt-jt-moderation-6b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 33.2, "gpqa_diamond": 25.0, "ifeval": 18.3, "math_500": 2.3, "mmlu_pro": 11.5, "musr": 34.6}, "capability_tiers": {}, "context_window": 2048, "cost_input": null, "display_name": "GPT NeoXT Chat Base 20B", "fits": {"amd_rx_7900_xt": 56.0, "amd_rx_7900_xtx": 67.2, "amd_rx_9070_xt": 45.1, "apple_m1_max": 28.0, "apple_m2_max": 28.0, "apple_m2_ultra": 56.0, "apple_m3_max": 28.0, "apple_m3_ultra": 57.3, "apple_m4": 8.4, "apple_m4_max": 38.2, "apple_m4_pro": 19.1, "nvidia_rtx_3090": 65.5, "nvidia_rtx_4060_ti_16gb": 20.2, "nvidia_rtx_4070_ti_super": 47.0, "nvidia_rtx_4080_super": 51.5, "nvidia_rtx_4090": 70.6, "nvidia_rtx_5080": 67.2, "nvidia_rtx_5090": 125.4}, "model_id": "together/gpt-neoxt-chat-base-20b", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 34.4, "gpqa_diamond": 25.2, "ifeval": 21.3, "math_500": 1.6, "mmlu_pro": 17.8, "musr": 40.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "Llama 2 7B 32K Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/llama-2-7b-32k-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 34.0, "gpqa_diamond": 25.0, "ifeval": 18.6, "math_500": 1.4, "mmlu_pro": 17.7, "musr": 37.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": null, "display_name": "LLaMA 2 7B 32K", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/llama-2-7b-32k", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 2048, "cost_input": null, "display_name": "Pythia Chat Base 7B", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/pythia-chat-base-7b", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "Together AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 31.9, "gpqa_diamond": 25.5, "ifeval": 20.8, "math_500": 1.6, "mmlu_pro": 12.0, "musr": 36.2}, "capability_tiers": {}, "context_window": 2048, "cost_input": null, "display_name": "RedPajama INCITE 7B Base", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/redpajama-incite-7b-base", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 31.8, "gpqa_diamond": 25.3, "ifeval": 15.6, "math_500": 0.7, "mmlu_pro": 11.2, "musr": 34.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "RedPajama INCITE 7B Chat", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/redpajama-incite-7b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 33.8, "gpqa_diamond": 25.1, "ifeval": 20.6, "math_500": 2.1, "mmlu_pro": 12.7, "musr": 36.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "RedPajama INCITE 7B Instruct", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "together/redpajama-incite-7b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 30.6, "gpqa_diamond": 24.3, "ifeval": 22.9, "math_500": 1.4, "mmlu_pro": 11.1, "musr": 37.4}, "capability_tiers": {}, "context_window": 2048, "cost_input": null, "display_name": "RedPajama INCITE Base 3B v1", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "together/redpajama-incite-base-3b-v1", "model_subtypes": [], "model_type": "llm-base", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 32.2, "gpqa_diamond": 24.4, "ifeval": 16.5, "math_500": 0.9, "mmlu_pro": 11.3, "musr": 36.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 2048, "cost_input": null, "display_name": "RedPajama INCITE Chat 3B v1", "fits": {"amd_rx_7900_xt": 373.3, "amd_rx_7900_xtx": 448.0, "amd_rx_9070_xt": 301.0, "apple_m1_max": 186.7, "apple_m2_max": 186.7, "apple_m2_ultra": 373.3, "apple_m3_max": 186.7, "apple_m3_ultra": 382.2, "apple_m4": 56.0, "apple_m4_max": 254.8, "apple_m4_pro": 127.4, "nvidia_rtx_3060_12gb": 168.0, "nvidia_rtx_3090": 436.8, "nvidia_rtx_4060_ti_16gb": 134.4, "nvidia_rtx_4070_ti_super": 313.6, "nvidia_rtx_4080_super": 343.5, "nvidia_rtx_4090": 470.4, "nvidia_rtx_5080": 448.0, "nvidia_rtx_5090": 836.3}, "model_id": "together/redpajama-incite-chat-3b-v1", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Together AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "DeepSeek OCR 2", "fits": {"apple_m3_ultra": 1.7}, "model_id": "unsloth/deepseek-ocr-2", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1290.0, "arena_elo_math": 1270.0, "arena_elo_overall": 1280.0, "gpqa_diamond": 65.2, "humaneval": 82.5, "ifeval": 81.5, "math_500": 85.1, "mmlu_pro": 72.3, "swe_bench_verified": 28.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.2, "display_name": "gemma 4 26B A4B it GGUF", "fits": {"amd_rx_7900_xt": 43.1, "amd_rx_7900_xtx": 51.7, "apple_m1_max": 21.5, "apple_m2_max": 21.5, "apple_m2_ultra": 43.1, "apple_m3_max": 21.5, "apple_m3_ultra": 44.1, "apple_m4": 6.5, "apple_m4_max": 29.4, "apple_m4_pro": 14.7, "nvidia_rtx_3090": 50.4, "nvidia_rtx_4090": 54.3, "nvidia_rtx_5090": 96.5}, "model_id": "unsloth/gemma-4-26b-a4b-it-gguf", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 54.8, "arena_elo_coding": 1300.0, "arena_elo_math": 1280.0, "arena_elo_overall": 1290.0, "artificial_analysis_quality_index": 73.0, "artificial_analysis_speed_index": 84.0, "bbh": 62.3, "chartqa": 78.5, "docvqa": 86.8, "gpqa_diamond": 67.8, "humaneval": 84.1, "ifeval": 83.0, "math_500": 87.3, "mathvista": 54.5, "mmlu_astronomy": 73.5, "mmlu_biology": 81.5, "mmlu_business_ethics": 74.2, "mmlu_chemistry": 74.5, "mmlu_clinical_knowledge": 80.2, "mmlu_computer_science": 82.5, "mmlu_jurisprudence": 71.5, "mmlu_physics": 78.8, "mmlu_pro": 74.1, "mmlu_professional_accounting": 65.2, "mmlu_professional_law": 70.2, "mmmu": 60.2, "multipl_e": 77.3, "multipl_e_cpp": 76.5, "multipl_e_csharp": 74.2, "multipl_e_go": 73.2, "multipl_e_java": 81.2, "multipl_e_javascript": 79.5, "multipl_e_julia": 54.2, "multipl_e_kotlin": 65.2, "multipl_e_lua": 47.5, "multipl_e_perl": 42.5, "multipl_e_php": 71.5, "multipl_e_python": 84.1, "multipl_e_r": 50.5, "multipl_e_ruby": 57.8, "multipl_e_rust": 67.5, "multipl_e_scala": 52.5, "multipl_e_swift": 57.8, "multipl_e_typescript": 78.8, "musr": 30.4, "swe_bench_verified": 30.2}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-1"}, "context_window": 131072, "cost_input": 0.2, "display_name": "gemma 4 31B it GGUF", "fits": {"amd_rx_7900_xtx": 43.4, "apple_m1_max": 18.1, "apple_m2_max": 18.1, "apple_m2_ultra": 36.1, "apple_m3_max": 18.1, "apple_m3_ultra": 37.0, "apple_m4": 5.4, "apple_m4_max": 24.7, "apple_m4_pro": 12.3, "nvidia_rtx_3090": 42.3, "nvidia_rtx_4090": 45.5, "nvidia_rtx_5090": 80.9}, "model_id": "unsloth/gemma-4-31b-it-gguf", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.05, "display_name": "gemma 4 E4B it GGUF", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "unsloth/gemma-4-e4b-it-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.7 Flash FP8 Dynamic", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "unsloth/glm-4-7-flash-fp8-dynamic", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "gpt oss 20B GGUF", "fits": {"amd_rx_7900_xt": 56.0, "amd_rx_7900_xtx": 67.2, "amd_rx_9070_xt": 45.1, "apple_m1_max": 28.0, "apple_m2_max": 28.0, "apple_m2_ultra": 56.0, "apple_m3_max": 28.0, "apple_m3_ultra": 57.3, "apple_m4": 8.4, "apple_m4_max": 38.2, "apple_m4_pro": 19.1, "nvidia_rtx_3090": 65.5, "nvidia_rtx_4060_ti_16gb": 20.2, "nvidia_rtx_4070_ti_super": 47.0, "nvidia_rtx_4080_super": 51.5, "nvidia_rtx_4090": 70.6, "nvidia_rtx_5080": 67.2, "nvidia_rtx_5090": 125.4}, "model_id": "unsloth/gpt-oss-20b-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1160.0, "arena_elo_math": 1150.0, "arena_elo_overall": 1170.0, "artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0, "bbh": 39.2, "bbq": 72.8, "gpqa_diamond": 28.5, "helm_safety": 78.5, "humaneval": 55.5, "ifeval": 62.0, "math_500": 42.0, "mmlu_pro": 40.5, "musr": 12.8, "toxigen": 82.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Llama 3.1 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "unsloth/llama-3-1-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "LTX 2.3 GGUF", "fits": {"amd_rx_7900_xt": 1120.0, "amd_rx_7900_xtx": 1344.0, "amd_rx_9070_xt": 903.0, "apple_m1_max": 560.0, "apple_m2_max": 560.0, "apple_m2_ultra": 1120.0, "apple_m3_max": 560.0, "apple_m3_ultra": 1146.6, "apple_m4": 168.0, "apple_m4_max": 764.4, "apple_m4_pro": 382.2, "nvidia_rtx_3060_12gb": 504.0, "nvidia_rtx_3090": 1310.4, "nvidia_rtx_4060_ti_16gb": 403.2, "nvidia_rtx_4070_ti_super": 940.8, "nvidia_rtx_4080_super": 1030.4, "nvidia_rtx_4090": 1411.2, "nvidia_rtx_5080": 1344.0, "nvidia_rtx_5090": 2508.8}, "model_id": "unsloth/ltx-2-3-gguf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Meta Llama 3.1 8B Instruct bnb 4bit", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "unsloth/meta-llama-3-1-8b-instruct-bnb-4bit", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {"artificial_analysis_quality_index": 62.0, "artificial_analysis_speed_index": 92.0}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.1, "display_name": "Meta Llama 3.1 8B Instruct", "fits": {"amd_rx_7900_xt": 140.0, "amd_rx_7900_xtx": 168.0, "amd_rx_9070_xt": 112.9, "apple_m1_max": 70.0, "apple_m2_max": 70.0, "apple_m2_ultra": 140.0, "apple_m3_max": 70.0, "apple_m3_ultra": 143.3, "apple_m4": 21.0, "apple_m4_max": 95.5, "apple_m4_pro": 47.8, "nvidia_rtx_3060_12gb": 63.0, "nvidia_rtx_3090": 163.8, "nvidia_rtx_4060_ti_16gb": 50.4, "nvidia_rtx_4070_ti_super": 117.6, "nvidia_rtx_4080_super": 128.8, "nvidia_rtx_4090": 176.4, "nvidia_rtx_5080": 168.0, "nvidia_rtx_5090": 313.6}, "model_id": "unsloth/meta-llama-3-1-8b-instruct", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 60.5, "bbh": 45.2, "gpqa_diamond": 29.2, "gsm8k": 34.5, "hellaswag": 83.0, "ifeval": 22.7, "math_500": 3.0, "mmlu_abstract_algebra": 30.0, "mmlu_anatomy": 60.7, "mmlu_astronomy": 65.1, "mmlu_business_ethics": 61.0, "mmlu_clinical_knowledge": 68.3, "mmlu_college_biology": 69.4, "mmlu_college_chemistry": 49.0, "mmlu_college_computer_science": 52.0, "mmlu_college_mathematics": 37.0, "mmlu_college_medicine": 63.0, "mmlu_college_physics": 31.4, "mmlu_computer_security": 73.0, "mmlu_conceptual_physics": 58.3, "mmlu_econometrics": 44.7, "mmlu_electrical_engineering": 55.9, "mmlu_elementary_mathematics": 40.2, "mmlu_formal_logic": 39.7, "mmlu_global_facts": 40.0, "mmlu_high_school_biology": 76.1, "mmlu_high_school_chemistry": 54.2, "mmlu_high_school_computer_science": 65.0, "mmlu_high_school_european_history": 76.4, "mmlu_high_school_geography": 77.3, "mmlu_high_school_government_and_politics": 87.6, "mmlu_high_school_macroeconomics": 60.5, "mmlu_high_school_mathematics": 35.9, "mmlu_high_school_microeconomics": 63.9, "mmlu_high_school_physics": 35.8, "mmlu_high_school_psychology": 80.0, "mmlu_high_school_statistics": 51.9, "mmlu_high_school_us_history": 81.4, "mmlu_high_school_world_history": 78.1, "mmlu_human_aging": 70.0, "mmlu_human_sexuality": 77.1, "mmlu_international_law": 81.0, "mmlu_jurisprudence": 74.1, "mmlu_logical_fallacies": 78.5, "mmlu_machine_learning": 53.6, "mmlu_management": 81.6, "mmlu_marketing": 88.0, "mmlu_medical_genetics": 70.0, "mmlu_miscellaneous": 79.8, "mmlu_moral_disputes": 70.2, "mmlu_moral_scenarios": 39.8, "mmlu_nutrition": 74.2, "mmlu_philosophy": 72.7, "mmlu_prehistory": 68.8, "mmlu_pro": 29.5, "mmlu_professional_accounting": 51.1, "mmlu_professional_law": 46.2, "mmlu_professional_medicine": 68.8, "mmlu_professional_psychology": 66.2, "mmlu_public_relations": 64.5, "mmlu_security_studies": 72.7, "mmlu_sociology": 82.6, "mmlu_us_foreign_policy": 88.0, "mmlu_virology": 53.6, "mmlu_world_religions": 81.9, "musr": 40.3, "truthfulqa": 41.8, "winogrande": 78.5}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.1, "display_name": "mistral 7B v0.3 bnb 4bit", "fits": {"amd_rx_7900_xt": 160.0, "amd_rx_7900_xtx": 192.0, "amd_rx_9070_xt": 129.0, "apple_m1_max": 80.0, "apple_m2_max": 80.0, "apple_m2_ultra": 160.0, "apple_m3_max": 80.0, "apple_m3_ultra": 163.8, "apple_m4": 24.0, "apple_m4_max": 109.2, "apple_m4_pro": 54.6, "nvidia_rtx_3060_12gb": 72.0, "nvidia_rtx_3090": 187.2, "nvidia_rtx_4060_ti_16gb": 57.6, "nvidia_rtx_4070_ti_super": 134.4, "nvidia_rtx_4080_super": 147.2, "nvidia_rtx_4090": 201.6, "nvidia_rtx_5080": 192.0, "nvidia_rtx_5090": 358.4}, "model_id": "unsloth/mistral-7b-v0-3-bnb-4bit", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.03, "display_name": "Qwen3 0.6B", "fits": {"amd_rx_7900_xt": 1866.7, "amd_rx_7900_xtx": 2240.0, "amd_rx_9070_xt": 1505.0, "apple_m1_max": 933.3, "apple_m2_max": 933.3, "apple_m2_ultra": 1866.7, "apple_m3_max": 933.3, "apple_m3_ultra": 1911.0, "apple_m4": 280.0, "apple_m4_max": 1274.0, "apple_m4_pro": 637.0, "nvidia_rtx_3060_12gb": 840.0, "nvidia_rtx_3090": 2184.0, "nvidia_rtx_4060_ti_16gb": 672.0, "nvidia_rtx_4070_ti_super": 1568.0, "nvidia_rtx_4080_super": 1717.3, "nvidia_rtx_4090": 2352.0, "nvidia_rtx_5080": 2240.0, "nvidia_rtx_5090": 4181.3}, "model_id": "unsloth/qwen3-0-6b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "Qwen3.5 0.8B GGUF", "fits": {"amd_rx_7900_xt": 1400.0, "amd_rx_7900_xtx": 1680.0, "amd_rx_9070_xt": 1128.7, "apple_m1_max": 700.0, "apple_m2_max": 700.0, "apple_m2_ultra": 1400.0, "apple_m3_max": 700.0, "apple_m3_ultra": 1433.2, "apple_m4": 210.0, "apple_m4_max": 955.5, "apple_m4_pro": 477.7, "nvidia_rtx_3060_12gb": 630.0, "nvidia_rtx_3090": 1638.0, "nvidia_rtx_4060_ti_16gb": 504.0, "nvidia_rtx_4070_ti_super": 1176.0, "nvidia_rtx_4080_super": 1288.0, "nvidia_rtx_4090": 1764.0, "nvidia_rtx_5080": 1680.0, "nvidia_rtx_5090": 3136.0}, "model_id": "unsloth/qwen3-5-0-8b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.55, "display_name": "Qwen3.5 122B A10B GGUF", "fits": {"apple_m2_max": 4.6, "apple_m2_ultra": 9.2, "apple_m3_max": 4.6, "apple_m3_ultra": 9.4, "apple_m4_max": 6.3}, "model_id": "unsloth/qwen3-5-122b-a10b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.25, "display_name": "Qwen3.5 27B GGUF", "fits": {"amd_rx_7900_xt": 41.5, "amd_rx_7900_xtx": 49.8, "apple_m1_max": 20.7, "apple_m2_max": 20.7, "apple_m2_ultra": 41.5, "apple_m3_max": 20.7, "apple_m3_ultra": 42.5, "apple_m4": 6.2, "apple_m4_max": 28.3, "apple_m4_pro": 14.2, "nvidia_rtx_3090": 48.5, "nvidia_rtx_4090": 52.3, "nvidia_rtx_5090": 92.9}, "model_id": "unsloth/qwen3-5-27b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.04, "display_name": "Qwen3.5 2B GGUF", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "unsloth/qwen3-5-2b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.15, "display_name": "Qwen3.5 35B A3B GGUF", "fits": {"amd_rx_7900_xtx": 38.4, "apple_m1_max": 16.0, "apple_m2_max": 16.0, "apple_m2_ultra": 32.0, "apple_m3_max": 16.0, "apple_m3_ultra": 32.8, "apple_m4": 4.8, "apple_m4_max": 21.8, "apple_m4_pro": 10.9, "nvidia_rtx_3090": 37.4, "nvidia_rtx_4090": 40.3, "nvidia_rtx_5090": 71.7}, "model_id": "unsloth/qwen3-5-35b-a3b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.05, "display_name": "Qwen3.5 4B GGUF", "fits": {"amd_rx_7900_xt": 280.0, "amd_rx_7900_xtx": 336.0, "amd_rx_9070_xt": 225.7, "apple_m1_max": 140.0, "apple_m2_max": 140.0, "apple_m2_ultra": 280.0, "apple_m3_max": 140.0, "apple_m3_ultra": 286.6, "apple_m4": 42.0, "apple_m4_max": 191.1, "apple_m4_pro": 95.5, "nvidia_rtx_3060_12gb": 126.0, "nvidia_rtx_3090": 327.6, "nvidia_rtx_4060_ti_16gb": 100.8, "nvidia_rtx_4070_ti_super": 235.2, "nvidia_rtx_4080_super": 257.6, "nvidia_rtx_4090": 352.8, "nvidia_rtx_5080": 336.0, "nvidia_rtx_5090": 627.2}, "model_id": "unsloth/qwen3-5-4b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": 0.1, "display_name": "Qwen3.5 9B GGUF", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "unsloth/qwen3-5-9b-gguf", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 262144, "cost_input": null, "display_name": "Qwen3 Coder Next GGUF", "fits": {"apple_m3_ultra": 2.4}, "model_id": "unsloth/qwen3-coder-next-gguf", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Unsloth", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 32768, "cost_input": 0.15, "display_name": "solar-mini", "fits": {}, "model_id": "upstage/solar-mini", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "Upstage", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 65536, "cost_input": 0.25, "display_name": "solar-pro2", "fits": {}, "model_id": "upstage/solar-pro2", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Upstage", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.25, "display_name": "solar-pro3", "fits": {}, "model_id": "upstage/solar-pro3", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "Upstage", "scores_as_of": null}, {"benchmark_scores": {"beir": 51.2, "miracl": 48.0, "mteb_classification": 67.8, "mteb_clustering": 46.2, "mteb_overall": 62.1, "mteb_pair_classification": 83.0, "mteb_reranking": 55.5, "mteb_retrieval": 58.2, "mteb_sts": 80.5, "mteb_summarization": 28.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Voyage 3 Lite", "fits": {}, "model_id": "voyage/voyage-3-lite", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Voyage AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 56.5, "miracl": 58.5, "mteb_classification": 71.8, "mteb_clustering": 51.5, "mteb_overall": 67.2, "mteb_pair_classification": 86.0, "mteb_reranking": 59.0, "mteb_retrieval": 64.5, "mteb_sts": 84.0, "mteb_summarization": 30.5}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Voyage 3", "fits": {}, "model_id": "voyage/voyage-3", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Voyage AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 52.5, "miracl": 42.0, "mteb_classification": 68.5, "mteb_clustering": 47.0, "mteb_overall": 63.5, "mteb_pair_classification": 83.5, "mteb_reranking": 56.5, "mteb_retrieval": 60.0, "mteb_sts": 81.0, "mteb_summarization": 29.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Voyage Code 3", "fits": {}, "model_id": "voyage/voyage-code-3", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Voyage AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 45.5, "mteb_classification": 68.8, "mteb_clustering": 46.8, "mteb_overall": 63.0, "mteb_pair_classification": 83.0, "mteb_reranking": 56.0, "mteb_retrieval": 59.8, "mteb_sts": 81.0, "mteb_summarization": 29.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Voyage Finance 2", "fits": {}, "model_id": "voyage/voyage-finance-2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Voyage AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"miracl": 44.0, "mteb_classification": 68.0, "mteb_clustering": 46.0, "mteb_overall": 62.5, "mteb_pair_classification": 82.8, "mteb_reranking": 55.8, "mteb_retrieval": 59.2, "mteb_sts": 80.8, "mteb_summarization": 28.8}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Voyage Law 2", "fits": {}, "model_id": "voyage/voyage-law-2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Voyage AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"beir": 52.0, "miracl": 57.0, "mteb_classification": 68.2, "mteb_clustering": 46.5, "mteb_overall": 62.8, "mteb_pair_classification": 83.5, "mteb_reranking": 56.0, "mteb_retrieval": 59.5, "mteb_sts": 81.5, "mteb_summarization": 29.0}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "Voyage Multilingual 2", "fits": {}, "model_id": "voyage/voyage-multilingual-2", "model_subtypes": [], "model_type": "embedding-text", "open_weights": false, "provider": "Voyage AI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 70.0, "gpqa_diamond": 50.1, "humaneval": 80.5, "ifeval": 78.0, "math_500": 72.0, "mmlu_pro": 65.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 2.0, "display_name": "Grok 2 (1212)", "fits": {}, "model_id": "xai/grok-2-1212", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 70.0, "chartqa": 73.2, "docvqa": 82.5, "gpqa_diamond": 50.1, "humaneval": 80.5, "ifeval": 78.0, "math_500": 72.0, "mathvista": 48.8, "mmlu_pro": 65.5, "mmmu": 55.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 2.0, "display_name": "Grok 2 Latest", "fits": {}, "model_id": "xai/grok-2-latest", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "chartqa": 73.2, "docvqa": 82.5, "gpqa_diamond": 50.1, "humaneval": 80.5, "ifeval": 78.0, "math_500": 72.0, "mathvista": 48.8, "mmlu_pro": 65.5, "mmmu": 55.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 2.0, "display_name": "Grok 2 Vision (1212)", "fits": {}, "model_id": "xai/grok-2-vision-1212", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "chartqa": 73.2, "docvqa": 82.5, "gpqa_diamond": 50.1, "humaneval": 80.5, "ifeval": 78.0, "math_500": 72.0, "mathvista": 48.8, "mmlu_pro": 65.5, "mmmu": 55.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 2.0, "display_name": "Grok 2 Vision Latest", "fits": {}, "model_id": "xai/grok-2-vision-latest", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "chartqa": 73.2, "docvqa": 82.5, "gpqa_diamond": 50.1, "humaneval": 80.5, "ifeval": 78.0, "math_500": 72.0, "mathvista": 48.8, "mmlu_pro": 65.5, "mmmu": 55.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 2.0, "display_name": "Grok 2 Vision", "fits": {}, "model_id": "xai/grok-2-vision", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"arena_elo_coding": 1260.0, "arena_elo_hard_prompts": 1325.7, "arena_elo_math": 1250.0, "arena_elo_overall": 1270.0, "arena_elo_style_control": 1334.8, "artificial_analysis_quality_index": 76.0, "artificial_analysis_speed_index": 70.0, "chartqa": 73.2, "docvqa": 82.5, "gpqa_diamond": 50.1, "humaneval": 80.5, "ifeval": 78.0, "math_500": 72.0, "mathvista": 48.8, "mmlu_pro": 65.5, "mmmu": 55.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 2.0, "display_name": "Grok 2", "fits": {}, "model_id": "xai/grok-2", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 53.3, "arena_elo_coding": 1320.0, "arena_elo_math": 1310.0, "arena_elo_overall": 1330.0, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 65.0, "gpqa_diamond": 62.8, "humaneval": 86.5, "ifeval": 85.0, "live_code_bench": 79.4, "math_500": 82.5, "mmlu_pro": 76.0, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5, "swe_bench_verified": 48.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 5.0, "display_name": "Grok 3 Fast Latest", "fits": {}, "model_id": "xai/grok-3-fast-latest", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 53.3, "arena_elo_coding": 1320.0, "arena_elo_math": 1310.0, "arena_elo_overall": 1330.0, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 65.0, "gpqa_diamond": 62.8, "humaneval": 86.5, "ifeval": 85.0, "live_code_bench": 79.4, "math_500": 82.5, "mmlu_pro": 76.0, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5, "swe_bench_verified": 48.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 5.0, "display_name": "Grok 3 Fast", "fits": {}, "model_id": "xai/grok-3-fast", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 53.3, "arena_elo_coding": 1320.0, "arena_elo_math": 1310.0, "arena_elo_overall": 1330.0, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 65.0, "gpqa_diamond": 62.8, "humaneval": 86.5, "ifeval": 85.0, "live_code_bench": 79.4, "math_500": 82.5, "mmlu_pro": 76.0, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5, "swe_bench_verified": 48.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 3.0, "display_name": "Grok 3 Latest", "fits": {}, "model_id": "xai/grok-3-latest", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 49.3, "arena_elo_coding": 1280.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1290.0, "artificial_analysis_quality_index": 72.0, "artificial_analysis_speed_index": 82.0, "gpqa_diamond": 55.3, "humaneval": 82.0, "ifeval": 80.5, "live_code_bench": 69.6, "math_500": 80.5, "mmlu_pro": 70.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.6, "display_name": "Grok 3 Mini Fast Latest", "fits": {}, "model_id": "xai/grok-3-mini-fast-latest", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 49.3, "arena_elo_coding": 1280.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1290.0, "artificial_analysis_quality_index": 72.0, "artificial_analysis_speed_index": 82.0, "gpqa_diamond": 55.3, "humaneval": 82.0, "ifeval": 80.5, "live_code_bench": 69.6, "math_500": 80.5, "mmlu_pro": 70.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.6, "display_name": "Grok 3 Mini Fast", "fits": {}, "model_id": "xai/grok-3-mini-fast", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 49.3, "arena_elo_coding": 1280.0, "arena_elo_math": 1300.0, "arena_elo_overall": 1290.0, "artificial_analysis_quality_index": 72.0, "artificial_analysis_speed_index": 82.0, "gpqa_diamond": 55.3, "humaneval": 82.0, "ifeval": 80.5, "live_code_bench": 69.6, "math_500": 80.5, "mmlu_pro": 70.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "Grok 3 Mini Latest", "fits": {}, "model_id": "xai/grok-3-mini-latest", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 49.3, "arena_elo_coding": 1280.0, "arena_elo_hard_prompts": 1377.8, "arena_elo_math": 1300.0, "arena_elo_overall": 1290.0, "arena_elo_style_control": 1362.7, "artificial_analysis_quality_index": 72.0, "artificial_analysis_speed_index": 82.0, "gpqa_diamond": 55.3, "humaneval": 82.0, "ifeval": 80.5, "live_code_bench": 69.6, "math_500": 80.5, "mmlu_pro": 70.2, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 0.3, "display_name": "Grok 3 Mini", "fits": {}, "model_id": "xai/grok-3-mini", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 53.3, "arena_elo_coding": 1320.0, "arena_elo_hard_prompts": 1426.2, "arena_elo_math": 1310.0, "arena_elo_overall": 1330.0, "arena_elo_style_control": 1411.7, "artificial_analysis_quality_index": 82.0, "artificial_analysis_speed_index": 65.0, "gpqa_diamond": 62.8, "humaneval": 86.5, "ifeval": 85.0, "live_code_bench": 79.4, "math_500": 82.5, "medqa": 86.1, "mmlu_pro": 76.0, "multipl_e_csharp": 78.5, "multipl_e_julia": 60.2, "multipl_e_kotlin": 71.2, "multipl_e_lua": 53.5, "multipl_e_perl": 49.2, "multipl_e_php": 76.2, "multipl_e_r": 56.8, "multipl_e_ruby": 64.5, "multipl_e_scala": 58.8, "multipl_e_swift": 64.5, "swe_bench_verified": 48.5}, "capability_tiers": {"coding": "tier-2", "reasoning": "tier-2", "reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 3.0, "display_name": "Grok 3", "fits": {}, "model_id": "xai/grok-3", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 81.9, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 2000000, "cost_input": 0.2, "display_name": "Grok 4.1 Fast (Non-Reasoning)", "fits": {}, "model_id": "xai/grok-4-1-fast-non-reasoning", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_hard_prompts": 1441.8, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "arena_elo_style_control": 1431.1, "arena_elo_vision": 1173.7, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 81.9, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 2000000, "cost_input": 0.2, "display_name": "Grok 4.1 Fast", "fits": {}, "model_id": "xai/grok-4-1-fast", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 81.9, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 2000000, "cost_input": 2.0, "display_name": "Grok 4.20 (Non-Reasoning)", "fits": {}, "model_id": "xai/grok-4-20-0309-non-reasoning", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_hard_prompts": 1505.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "arena_elo_style_control": 1490.4, "arena_elo_vision": 1242.8, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 81.9, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 2000000, "cost_input": 2.0, "display_name": "Grok 4.20 (Reasoning)", "fits": {}, "model_id": "xai/grok-4-20-0309-reasoning", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_hard_prompts": 1491.6, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "arena_elo_style_control": 1474.5, "arena_elo_vision": 1228.4, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 81.9, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1"}, "context_window": 2000000, "cost_input": 2.0, "display_name": "Grok 4.20 Multi-Agent", "fits": {}, "model_id": "xai/grok-4-20-multi-agent-0309", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_hard_prompts": 1432.7, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "arena_elo_style_control": 1420.7, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 83.2, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 2000000, "cost_input": 0.2, "display_name": "Grok 4 Fast (Non-Reasoning)", "fits": {}, "model_id": "xai/grok-4-fast-non-reasoning", "model_subtypes": ["llm-code", "llm-reasoning"], "model_type": "vlm", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 83.2, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 2000000, "cost_input": 0.2, "display_name": "Grok 4 Fast", "fits": {}, "model_id": "xai/grok-4-fast", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {"aider_polyglot": 70.8, "arena_elo_coding": 1360.0, "arena_elo_hard_prompts": 1482.8, "arena_elo_math": 1380.0, "arena_elo_overall": 1370.0, "arena_elo_style_control": 1470.7, "arena_elo_vision": 1182.0, "gpqa_diamond": 75.2, "humaneval": 91.0, "ifeval": 90.0, "live_code_bench": 81.9, "math_500": 94.0, "mmlu_pro": 82.5, "multipl_e_csharp": 85.2, "multipl_e_julia": 67.8, "multipl_e_kotlin": 79.5, "multipl_e_lua": 61.2, "multipl_e_perl": 57.5, "multipl_e_php": 82.8, "multipl_e_r": 64.5, "multipl_e_ruby": 73.5, "multipl_e_scala": 68.2, "multipl_e_swift": 72.8, "swe_bench_agent": 50.2, "swe_bench_verified": 64.5}, "capability_tiers": {"coding": "tier-1", "reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "reasoning:mathematical": "tier-1", "reasoning:multi_step": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 256000, "cost_input": 3.0, "display_name": "Grok 4", "fits": {}, "model_id": "xai/grok-4", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": 5.0, "display_name": "Grok Beta", "fits": {}, "model_id": "xai/grok-beta", "model_subtypes": [], "model_type": "llm-chat", "open_weights": false, "provider": "xAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 256000, "cost_input": 0.2, "display_name": "Grok Code Fast 1", "fits": {}, "model_id": "xai/grok-code-fast-1", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": false, "provider": "xAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": 5.0, "display_name": "Grok Vision Beta", "fits": {}, "model_id": "xai/grok-vision-beta", "model_subtypes": [], "model_type": "vlm", "open_weights": false, "provider": "xAI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "AutoGLM Phone 9B Multilingual", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/autoglm-phone-9b-multilingual", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "AutoGLM Phone 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/autoglm-phone-9b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arc_challenge": 38.8, "hellaswag": 59.0, "mmlu_abstract_algebra": 28.0, "mmlu_anatomy": 41.5, "mmlu_astronomy": 53.3, "mmlu_business_ethics": 61.0, "mmlu_clinical_knowledge": 48.7, "mmlu_college_biology": 38.9, "mmlu_college_chemistry": 40.0, "mmlu_college_computer_science": 33.0, "mmlu_college_mathematics": 32.0, "mmlu_college_medicine": 42.2, "mmlu_college_physics": 21.6, "mmlu_computer_security": 56.0, "mmlu_conceptual_physics": 40.0, "mmlu_econometrics": 30.7, "mmlu_electrical_engineering": 46.9, "mmlu_elementary_mathematics": 31.7, "mmlu_formal_logic": 34.9, "mmlu_global_facts": 36.0, "mmlu_high_school_biology": 55.8, "mmlu_high_school_chemistry": 45.3, "mmlu_high_school_computer_science": 42.0, "mmlu_high_school_european_history": 60.0, "mmlu_high_school_geography": 55.6, "mmlu_high_school_government_and_politics": 57.0, "mmlu_high_school_macroeconomics": 44.9, "mmlu_high_school_mathematics": 26.3, "mmlu_high_school_microeconomics": 43.3, "mmlu_high_school_physics": 29.1, "mmlu_high_school_psychology": 59.3, "mmlu_high_school_statistics": 33.8, "mmlu_high_school_us_history": 58.8, "mmlu_high_school_world_history": 66.2, "mmlu_human_aging": 46.6, "mmlu_human_sexuality": 48.1, "mmlu_international_law": 60.3, "mmlu_jurisprudence": 58.3, "mmlu_logical_fallacies": 49.1, "mmlu_machine_learning": 43.8, "mmlu_management": 62.1, "mmlu_marketing": 69.2, "mmlu_medical_genetics": 52.0, "mmlu_miscellaneous": 59.6, "mmlu_moral_disputes": 52.9, "mmlu_moral_scenarios": 23.6, "mmlu_nutrition": 53.3, "mmlu_philosophy": 49.8, "mmlu_prehistory": 49.4, "mmlu_professional_accounting": 33.3, "mmlu_professional_law": 35.6, "mmlu_professional_medicine": 35.3, "mmlu_professional_psychology": 44.4, "mmlu_public_relations": 58.2, "mmlu_security_studies": 56.7, "mmlu_sociology": 62.2, "mmlu_us_foreign_policy": 73.0, "mmlu_virology": 41.6, "mmlu_world_religions": 47.4, "truthfulqa": 48.1}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "chatglm2 6B", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "zhipu/chatglm2-6b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": "2024-07"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "chatglm3 6B base", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "zhipu/chatglm3-6b-base", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "chatglm3 6B", "fits": {"amd_rx_7900_xt": 186.7, "amd_rx_7900_xtx": 224.0, "amd_rx_9070_xt": 150.5, "apple_m1_max": 93.3, "apple_m2_max": 93.3, "apple_m2_ultra": 186.7, "apple_m3_max": 93.3, "apple_m3_ultra": 191.1, "apple_m4": 28.0, "apple_m4_max": 127.4, "apple_m4_pro": 63.7, "nvidia_rtx_3060_12gb": 84.0, "nvidia_rtx_3090": 218.4, "nvidia_rtx_4060_ti_16gb": 67.2, "nvidia_rtx_4070_ti_super": 156.8, "nvidia_rtx_4080_super": 171.7, "nvidia_rtx_4090": 235.2, "nvidia_rtx_5080": 224.0, "nvidia_rtx_5090": 418.1}, "model_id": "zhipu/chatglm3-6b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "codegeex4 all 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/codegeex4-all-9b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "CogVideoX 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "zhipu/cogvideox-2b", "model_subtypes": [], "model_type": "video-generation", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "CogVideoX 5B I2V", "fits": {"amd_rx_7900_xt": 224.0, "amd_rx_7900_xtx": 268.8, "amd_rx_9070_xt": 180.6, "apple_m1_max": 112.0, "apple_m2_max": 112.0, "apple_m2_ultra": 224.0, "apple_m3_max": 112.0, "apple_m3_ultra": 229.3, "apple_m4": 33.6, "apple_m4_max": 152.9, "apple_m4_pro": 76.4, "nvidia_rtx_3060_12gb": 100.8, "nvidia_rtx_3090": 262.1, "nvidia_rtx_4060_ti_16gb": 80.6, "nvidia_rtx_4070_ti_super": 188.2, "nvidia_rtx_4080_super": 206.1, "nvidia_rtx_4090": 282.2, "nvidia_rtx_5080": 268.8, "nvidia_rtx_5090": 501.8}, "model_id": "zhipu/cogvideox-5b-i2v", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "CogVideoX 5B", "fits": {"amd_rx_7900_xt": 224.0, "amd_rx_7900_xtx": 268.8, "amd_rx_9070_xt": 180.6, "apple_m1_max": 112.0, "apple_m2_max": 112.0, "apple_m2_ultra": 224.0, "apple_m3_max": 112.0, "apple_m3_ultra": 229.3, "apple_m4": 33.6, "apple_m4_max": 152.9, "apple_m4_pro": 76.4, "nvidia_rtx_3060_12gb": 100.8, "nvidia_rtx_3090": 262.1, "nvidia_rtx_4060_ti_16gb": 80.6, "nvidia_rtx_4070_ti_super": 188.2, "nvidia_rtx_4080_super": 206.1, "nvidia_rtx_4090": 282.2, "nvidia_rtx_5080": 268.8, "nvidia_rtx_5090": 501.8}, "model_id": "zhipu/cogvideox-5b", "model_subtypes": [], "model_type": "video-generation", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": null, "cost_input": null, "display_name": "CogVideoX1.5 5B I2V", "fits": {"amd_rx_7900_xt": 224.0, "amd_rx_7900_xtx": 268.8, "amd_rx_9070_xt": 180.6, "apple_m1_max": 112.0, "apple_m2_max": 112.0, "apple_m2_ultra": 224.0, "apple_m3_max": 112.0, "apple_m3_ultra": 229.3, "apple_m4": 33.6, "apple_m4_max": 152.9, "apple_m4_pro": 76.4, "nvidia_rtx_3060_12gb": 100.8, "nvidia_rtx_3090": 262.1, "nvidia_rtx_4060_ti_16gb": 80.6, "nvidia_rtx_4070_ti_super": 188.2, "nvidia_rtx_4080_super": 206.1, "nvidia_rtx_4090": 282.2, "nvidia_rtx_5080": 268.8, "nvidia_rtx_5090": 501.8}, "model_id": "zhipu/cogvideox1-5-5b-i2v", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "cogvlm2 llama3 chat 19B", "fits": {"amd_rx_7900_xt": 58.9, "amd_rx_7900_xtx": 70.7, "amd_rx_9070_xt": 47.5, "apple_m1_max": 29.5, "apple_m2_max": 29.5, "apple_m2_ultra": 58.9, "apple_m3_max": 29.5, "apple_m3_ultra": 60.3, "apple_m4": 8.8, "apple_m4_max": 40.2, "apple_m4_pro": 20.1, "nvidia_rtx_3090": 69.0, "nvidia_rtx_4060_ti_16gb": 21.2, "nvidia_rtx_4070_ti_super": 49.5, "nvidia_rtx_4080_super": 54.2, "nvidia_rtx_4090": 74.3, "nvidia_rtx_5080": 70.7, "nvidia_rtx_5090": 132.0}, "model_id": "zhipu/cogvlm2-llama3-chat-19b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "reasoning:mathematical": "tier-2", "reasoning:multi_step": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.1V 9B Thinking", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-1v-9b-thinking", "model_subtypes": [], "model_type": "llm-reasoning", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.5 Air FP8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-5-air-fp8", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1426.9, "arena_elo_hard_prompts": 1391.2, "arena_elo_math": 1390.0, "arena_elo_overall": 1372.8, "arena_elo_style_control": 1372.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.5 Air", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-5-air", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1454.6, "arena_elo_hard_prompts": 1432.7, "arena_elo_math": 1415.0, "arena_elo_overall": 1410.9, "arena_elo_style_control": 1410.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.5", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-5", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.5V FP8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-5v-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1403.8, "arena_elo_hard_prompts": 1375.5, "arena_elo_math": 1357.7, "arena_elo_overall": 1353.3, "arena_elo_style_control": 1353.3, "arena_elo_vision": 1156.6}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.5V", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-5v", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.6 FP8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-6-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1459.8, "arena_elo_hard_prompts": 1442.6, "arena_elo_math": 1421.6, "arena_elo_overall": 1425.8, "arena_elo_style_control": 1425.8}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.6", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-6", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.6V Flash", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-6v-flash", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.6V FP8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-6v-fp8", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1420.5, "arena_elo_hard_prompts": 1384.1, "arena_elo_overall": 1377.9, "arena_elo_style_control": 1377.9, "arena_elo_vision": 1163.1}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.6V", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-6v", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1425.0, "arena_elo_hard_prompts": 1387.9, "arena_elo_math": 1367.8, "arena_elo_overall": 1368.7, "arena_elo_style_control": 1368.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.7 Flash", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-7-flash", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.7 FP8", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-7-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1486.1, "arena_elo_hard_prompts": 1463.5, "arena_elo_math": 1429.7, "arena_elo_overall": 1442.7, "arena_elo_style_control": 1442.7}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4.7", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-7", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"bbh": 55.3, "gpqa_diamond": 31.6, "ifeval": 14.3, "math_500": 0.0, "mmlu_pro": 41.4, "musr": 43.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 4 9B 0414", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-9b-0414", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 41.8, "gpqa_diamond": 30.4, "ifeval": 0.0, "math_500": 0.0, "mmlu_pro": 31.6, "musr": 37.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 1048576, "cost_input": null, "display_name": "glm 4 9B chat 1M", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-9b-chat-1m", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 44.3, "gpqa_diamond": 30.3, "ifeval": 65.1, "math_500": 8.5, "mmlu_pro": 27.7, "musr": 35.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "glm 4 9B chat hf", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-9b-chat-hf", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 47.4, "gpqa_diamond": 31.4, "ifeval": 0.0, "math_500": 0.0, "mmlu_pro": 31.7, "musr": 39.9}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "glm 4 9B chat", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-9b-chat", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {"bbh": 55.3, "gpqa_diamond": 31.6, "ifeval": 14.3, "math_500": 0.0, "mmlu_pro": 41.4, "musr": 43.9}, "capability_tiers": {"coding": "tier-2", "coding:code_completion": "tier-2", "coding:debugging": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "glm 4 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-9b", "model_subtypes": [], "model_type": "llm-code", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": "2025-03"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "glm 4 voice 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4-voice-9b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 8192, "cost_input": null, "display_name": "glm 4 voice tokenizer", "fits": {}, "model_id": "zhipu/glm-4-voice-tokenizer", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "glm 4v 9B", "fits": {"amd_rx_7900_xt": 124.4, "amd_rx_7900_xtx": 149.3, "amd_rx_9070_xt": 100.3, "apple_m1_max": 62.2, "apple_m2_max": 62.2, "apple_m2_ultra": 124.4, "apple_m3_max": 62.2, "apple_m3_ultra": 127.4, "apple_m4": 18.7, "apple_m4_max": 84.9, "apple_m4_pro": 42.5, "nvidia_rtx_3060_12gb": 56.0, "nvidia_rtx_3090": 145.6, "nvidia_rtx_4060_ti_16gb": 44.8, "nvidia_rtx_4070_ti_super": 104.5, "nvidia_rtx_4080_super": 114.5, "nvidia_rtx_4090": 156.8, "nvidia_rtx_5080": 149.3, "nvidia_rtx_5090": 278.8}, "model_id": "zhipu/glm-4v-9b", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"aime_2025": 95.3, "arena_elo_coding": 1520.3, "arena_elo_hard_prompts": 1494.3, "arena_elo_math": 1457.4, "arena_elo_overall": 1467.4, "arena_elo_style_control": 1467.4, "browsecomp": 68.0, "gpqa_diamond": 86.2, "hle": 31.0, "hle_tools": 52.3, "swe_bench_pro": 58.4, "tau_bench": 70.6, "terminal_bench_2": 63.5, "usamo_2026": 83.8}, "capability_tiers": {"reasoning": "tier-1", "reasoning:chain_of_thought": "tier-1", "tool_use": "tier-1", "tool_use:function_calling": "tier-1"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 5.1", "fits": {"apple_m3_ultra": 1.5}, "model_id": "zhipu/glm-5-1", "model_subtypes": ["llm-code"], "model_type": "llm-reasoning", "open_weights": true, "provider": "Z.ai (Zhipu AI)", "scores_as_of": "2026-04"}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 5 FP8", "fits": {"apple_m3_ultra": 1.5}, "model_id": "zhipu/glm-5-fp8", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {"arena_elo_coding": 1488.6, "arena_elo_hard_prompts": 1475.9, "arena_elo_math": 1458.3, "arena_elo_overall": 1455.6, "arena_elo_style_control": 1455.6}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM 5", "fits": {"apple_m3_ultra": 1.5}, "model_id": "zhipu/glm-5", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "GLM ASR Nano 2512", "fits": {}, "model_id": "zhipu/glm-asr-nano-2512", "model_subtypes": [], "model_type": "audio-asr", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": 131072, "cost_input": null, "display_name": "glm edge v 2B", "fits": {"amd_rx_7900_xt": 560.0, "amd_rx_7900_xtx": 672.0, "amd_rx_9070_xt": 451.5, "apple_m1_max": 280.0, "apple_m2_max": 280.0, "apple_m2_ultra": 560.0, "apple_m3_max": 280.0, "apple_m3_ultra": 573.3, "apple_m4": 84.0, "apple_m4_max": 382.2, "apple_m4_pro": 191.1, "nvidia_rtx_3060_12gb": 252.0, "nvidia_rtx_3090": 655.2, "nvidia_rtx_4060_ti_16gb": 201.6, "nvidia_rtx_4070_ti_super": 470.4, "nvidia_rtx_4080_super": 515.2, "nvidia_rtx_4090": 705.6, "nvidia_rtx_5080": 672.0, "nvidia_rtx_5090": 1254.4}, "model_id": "zhipu/glm-edge-v-2b", "model_subtypes": [], "model_type": "vlm", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "GLM Image", "fits": {}, "model_id": "zhipu/glm-image", "model_subtypes": [], "model_type": "image-generation", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {}, "context_window": null, "cost_input": null, "display_name": "GLM OCR", "fits": {}, "model_id": "zhipu/glm-ocr", "model_subtypes": [], "model_type": "document-ocr", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}, {"benchmark_scores": {}, "capability_tiers": {"reasoning:chain_of_thought": "tier-2", "tool_use": "tier-2", "tool_use:function_calling": "tier-2"}, "context_window": 131072, "cost_input": null, "display_name": "GLM Z1 32B 0414", "fits": {"amd_rx_7900_xtx": 42.0, "apple_m1_max": 17.5, "apple_m2_max": 17.5, "apple_m2_ultra": 35.0, "apple_m3_max": 17.5, "apple_m3_ultra": 35.8, "apple_m4": 5.2, "apple_m4_max": 23.9, "apple_m4_pro": 11.9, "nvidia_rtx_3090": 40.9, "nvidia_rtx_4090": 44.1, "nvidia_rtx_5090": 78.4}, "model_id": "zhipu/glm-z1-32b-0414", "model_subtypes": [], "model_type": "llm-chat", "open_weights": true, "provider": "Zhipu AI", "scores_as_of": null}], "count": 1143}