DeepSeek R1 0528 NVFP4 v2

NVIDIA · nvidia/deepseek-r1-0528-nvfp4-v2

ProviderNVIDIA
Familynvidia
Typellm-code
Statusactive
Released2025-07-21

Reported benchmark scores

These scores carry one collection date for the whole card (2026-04) and a source list rather than a source per score, so they cannot be attributed individually. They are shown as reported and are not verified evidence. Benchmark pages carry the reviewed, dated evidence where it exists.
BenchmarkCatalogue standingScore
Aider Polyglot Benchmarkunassessed65.8
AIME 2025unassessed82.5
AlpacaEvalunassessed45.2
Arena Elo — Codingunassessed1380.0
Arena Elo — Mathunassessed1410.0
Arena Elo — Overall (Text)unassessed1370.0
Artificial Analysis Intelligence Indexunassessed84.0
Artificial Analysis Output Speedunassessed40.0
BIG-Bench Hardunassessed68.7
GPQA Diamondunassessed73.2
HumanEvalunassessed90.5
IFEvalunassessed88.0
LiveCodeBenchunassessed77.0
MATH-500unassessed97.5
MMLU: Astronomyunassessed76.5
MMLU: Biology (subcategory)unassessed84.2
MMLU: Business Ethicsunassessed77.8
MMLU: Chemistry (subcategory)unassessed78.5
MMLU: Clinical Knowledgeunassessed83.5
MMLU: Computer Science (subcategory)unassessed85.5
MMLU: Jurisprudenceunassessed74.5
MMLU: Physics (unresolved key)unassessed82.8
MMLU-Prounassessed81.0
MMLU: Professional Accountingunassessed68.2
MMLU: Professional Lawunassessed72.5
MT-Benchunassessed9.0
MultiPL-Eunassessed84.6
MultiPL-E: C++unassessed83.5
MultiPL-E: C#unassessed82.1
MultiPL-E: Gounassessed80.5
MultiPL-E: Javaunassessed87.2
MultiPL-E: JavaScriptunassessed86.8
MultiPL-E: Juliaunassessed64.5
MultiPL-E: Kotlin (unconfirmed)unassessed74.8
MultiPL-E: Luaunassessed57.8
MultiPL-E: Perlunassessed52.5
MultiPL-E: PHPunassessed79.5
MultiPL-E: Pythonunassessed92.8
MultiPL-E: Runassessed60.2
MultiPL-E: Rubyunassessed67.2
MultiPL-E: Rustunassessed75.1
MultiPL-E: Scalaunassessed62.5
MultiPL-E: Swiftunassessed67.5
MultiPL-E: TypeScriptunassessed86.1
MuSRunassessed33.5
SWE-bench Agentunassessed42.5
SWE-bench Verifiedunassessed53.8
τ-benchunassessed48.8
Terminal-Benchunassessed5.7
WildBenchunassessed72.5

Data

This card as JSON · Edit on GitHub