Essential AI · essentialai/rnj-1-instruct
| Provider | Essential AI |
|---|---|
| Family | rnj |
| Type | llm-code |
| Status | active |
| Released | 2025-12-08 |
| Updated | 2025-12-24 |
| Parameters | 8.3B |
| Open weights | yes |
| Relationship | Model |
|---|---|
| Is a finetune of | EssentialAI/rnj-1 |
Function Calling · tier-2
Ordered by predicted speed. Bandwidth sets decode rate; memory decides whether it runs at all.
| Device | Memory | Bandwidth | Best quality | Weights | tok/s | Fastest |
|---|---|---|---|---|---|---|
| Cerebras WSE-3 | 44.0 GB | 21000000.0 GB/s | bf16 | 16.62 GB | ~884423.2 | ~3537692.7 q4 |
| NVIDIA GB200 Grace Blackwell Superchip | 372.0 GB | 16000.0 GB/s | bf16 | 16.62 GB | ~673.8 | ~2695.4 q4 |
| AMD Instinct MI355X | 288.0 GB | 8000.0 GB/s | bf16 | 16.62 GB | ~336.9 | ~1347.7 q4 |
| NVIDIA B200 | 180.0 GB | 7700.0 GB/s | bf16 | 16.62 GB | ~324.3 | ~1297.2 q4 |
| Google TPU7x (Ironwood) | 192.0 GB | 7380.0 GB/s | bf16 | 16.62 GB | ~310.8 | ~1243.2 q4 |
| AMD Instinct MI325X | 256.0 GB | 6000.0 GB/s | bf16 | 16.62 GB | ~252.7 | ~1010.8 q4 |
| AMD Instinct MI300X | 192.0 GB | 5300.0 GB/s | bf16 | 16.62 GB | ~223.2 | ~892.8 q4 |
| NVIDIA H200 SXM | 141.0 GB | 4800.0 GB/s | bf16 | 16.62 GB | ~202.2 | ~808.6 q4 |
| NVIDIA H100 NVL | 94.0 GB | 3900.0 GB/s | bf16 | 16.62 GB | ~164.3 | ~657.0 q4 |
| NVIDIA H100 SXM | 80.0 GB | 3350.0 GB/s | bf16 | 16.62 GB | ~141.1 | ~564.3 q4 |
| AMD Instinct MI250X | 128.0 GB | 3200.0 GB/s | bf16 | 16.62 GB | ~134.8 | ~539.1 q4 |
| Google TPU v5p | 95.0 GB | 2765.0 GB/s | bf16 | 16.62 GB | ~116.4 | ~465.8 q4 |
| NVIDIA A100 80GB SXM | 80.0 GB | 2039.0 GB/s | bf16 | 16.62 GB | ~85.9 | ~343.5 q4 |
| NVIDIA H100 PCIe | 80.0 GB | 2000.0 GB/s | bf16 | 16.62 GB | ~84.2 | ~336.9 q4 |
| NVIDIA GeForce RTX 5090 | 32.0 GB | 1792.0 GB/s | bf16 | 16.62 GB | ~75.5 | ~301.9 q4 |
| Google TPU v6e (Trillium) | 32.0 GB | 1638.0 GB/s | bf16 | 16.62 GB | ~69.0 | ~275.9 q4 |
| AMD Instinct MI210 | 64.0 GB | 1600.0 GB/s | bf16 | 16.62 GB | ~67.4 | ~269.5 q4 |
| NVIDIA A100 40GB SXM | 40.0 GB | 1555.0 GB/s | bf16 | 16.62 GB | ~65.5 | ~262.0 q4 |
| Google TPU v4 | 32.0 GB | 1200.0 GB/s | bf16 | 16.62 GB | ~50.5 | ~202.2 q4 |
| NVIDIA GeForce RTX 4090 | 24.0 GB | 1008.0 GB/s | bf16 | 16.62 GB | ~42.5 | ~169.8 q4 |
| AMD Radeon RX 7900 XTX | 24.0 GB | 960.0 GB/s | bf16 | 16.62 GB | ~40.4 | ~161.7 q4 |
| NVIDIA GeForce RTX 5080 | 16.0 GB | 960.0 GB/s | fp8 | 8.31 GB | ~80.9 | ~161.7 q4 |
| NVIDIA GeForce RTX 3090 | 24.0 GB | 936.0 GB/s | bf16 | 16.62 GB | ~39.4 | ~157.7 q4 |
| NVIDIA L40S | 48.0 GB | 864.0 GB/s | bf16 | 16.62 GB | ~36.4 | ~145.6 q4 |
| Apple M3 Ultra | 512.0 GB | 819.0 GB/s | bf16 | 16.62 GB | ~34.5 | ~138.0 q4 |
| AMD Radeon RX 7900 XT | 20.0 GB | 800.0 GB/s | fp8 | 8.31 GB | ~67.4 | ~134.8 q4 |
| Apple M2 Ultra | 192.0 GB | 800.0 GB/s | bf16 | 16.62 GB | ~33.7 | ~134.8 q4 |
| Google TPU v5e | 16.0 GB | 800.0 GB/s | fp8 | 8.31 GB | ~67.4 | ~134.8 q4 |
| NVIDIA GeForce RTX 4080 SUPER | 16.0 GB | 736.0 GB/s | fp8 | 8.31 GB | ~62.0 | ~124.0 q4 |
| NVIDIA GeForce RTX 4070 Ti SUPER | 16.0 GB | 672.0 GB/s | fp8 | 8.31 GB | ~56.6 | ~113.2 q4 |
| AMD Radeon RX 9070 XT | 16.0 GB | 640.0 GB/s | fp8 | 8.31 GB | ~53.9 | ~107.8 q4 |
| Apple M4 Max | 128.0 GB | 546.0 GB/s | bf16 | 16.62 GB | ~23.0 | ~92.0 q4 |
| Apple M1 Max | 64.0 GB | 400.0 GB/s | bf16 | 16.62 GB | ~16.8 | ~67.4 q4 |
| Apple M2 Max | 96.0 GB | 400.0 GB/s | bf16 | 16.62 GB | ~16.8 | ~67.4 q4 |
| Apple M3 Max | 128.0 GB | 400.0 GB/s | bf16 | 16.62 GB | ~16.8 | ~67.4 q4 |
| NVIDIA GeForce RTX 3060 12GB | 12.0 GB | 360.0 GB/s | fp8 | 8.31 GB | ~30.3 | ~60.6 q4 |
| NVIDIA L4 | 24.0 GB | 300.0 GB/s | bf16 | 16.62 GB | ~12.6 | ~50.5 q4 |
| NVIDIA GeForce RTX 4060 Ti 16GB | 16.0 GB | 288.0 GB/s | fp8 | 8.31 GB | ~24.3 | ~48.5 q4 |
| Apple M4 Pro | 64.0 GB | 273.0 GB/s | bf16 | 16.62 GB | ~11.5 | ~46.0 q4 |
| Apple M4 | 32.0 GB | 120.0 GB/s | bf16 | 16.62 GB | ~5.1 | ~20.2 q4 |
Taken from the card's availability section.
| Platform | Id on that platform |
|---|---|
| Hugging Face | EssentialAI/rnj-1-instruct |
| LM Studio | rnj-1 |
| Ollama | rnj-1 |
| OpenRouter | essentialai/rnj-1-instruct |
| Together AI | essentialai/rnj-1-instruct |
This card reports no benchmark scores yet.
| Benchmark | Catalogue standing | Score |
|---|