Mistral Nemo 12B hardware requirements
12.2B parameters. Needs about 9.3 GB at Q4_K_M with 8K context. Model card.
Memory needed at each quantization
| Quant | Quality | 8K context | 32K context |
|---|---|---|---|
| FP16 | Full quality | 26.3 GB | 30.4 GB |
| Q8_0 | Near-lossless | 14.9 GB | 18.9 GB |
| Q6_K | Excellent | 11.9 GB | 16 GB |
| Q5_K_M | Very good | 10.6 GB | 14.6 GB |
| Q4_K_M | Good, most popular | 9.3 GB | 13.4 GB |
| Q3_K_M | Noticeably degraded | 7.9 GB | 11.9 GB |
| Q2_K | Heavily degraded | 6.5 GB | 10.5 GB |
Run it with Ollama: ollama run mistral-nemo:12b (downloads Q4_K_M by default).
Run it in the cloud
Don't want to buy hardware? Run Mistral Nemo 12B on a rented GPU today.
Rent a cloud GPU by the hour. No upfront cost, cancel any time.
Every GPU and Mac checked
| GPU | Runs? | Best fit |
|---|---|---|
| NVIDIA RTX 3060 12GB | Yes | Q6_K · ~11.9 GB · ~22 tok/s |
| NVIDIA RTX 3070 | Yes, with heavy quantization | Q3_K_M · ~7.9 GB · ~45 tok/s |
| NVIDIA RTX 3080 10GB | Yes | Q4_K_M · ~9.3 GB · ~62 tok/s |
| NVIDIA RTX 3090 | Yes | Q8_0 · ~14.9 GB · ~43 tok/s |
| NVIDIA RTX 4060 | Yes, with heavy quantization | Q3_K_M · ~7.9 GB · ~27 tok/s |
| NVIDIA RTX 4060 Ti 16GB | Yes | Q8_0 · ~14.9 GB · ~13 tok/s |
| NVIDIA RTX 5060 Ti 16GB | Yes | Q8_0 · ~14.9 GB · ~21 tok/s |
| NVIDIA RTX 4070 | Yes | Q6_K · ~11.9 GB · ~30 tok/s |
| NVIDIA RTX 4070 Ti Super | Yes | Q8_0 · ~14.9 GB · ~31 tok/s |
| NVIDIA RTX 4080 | Yes | Q8_0 · ~14.9 GB · ~33 tok/s |
| NVIDIA RTX 4090 | Yes | Q8_0 · ~14.9 GB · ~47 tok/s |
| NVIDIA RTX 5070 | Yes | Q6_K · ~11.9 GB · ~40 tok/s |
| NVIDIA RTX 5070 Ti | Yes | Q8_0 · ~14.9 GB · ~41 tok/s |
| NVIDIA RTX 5080 | Yes | Q8_0 · ~14.9 GB · ~44 tok/s |
| NVIDIA RTX 5090 | Yes | FP16 · ~26.3 GB · ~44 tok/s |
| AMD Radeon RX 7800 XT | Yes | Q8_0 · ~14.9 GB · ~29 tok/s |
| AMD Radeon RX 7900 XTX | Yes | Q8_0 · ~14.9 GB · ~44 tok/s |
| Apple M4 (16GB) | Yes | Q5_K_M · ~10.6 GB · ~8 tok/s |
| Apple M1 (16GB) | Yes | Q5_K_M · ~10.6 GB · ~5 tok/s |
| Apple M2 Pro (32GB) | Yes | Q8_0 · ~14.9 GB · ~9 tok/s |
| Apple M3 Max (64GB) | Yes | FP16 · ~26.3 GB · ~10 tok/s |
| Apple M4 Pro (48GB) | Yes | FP16 · ~26.3 GB · ~7 tok/s |
| Apple M4 Max (128GB) | Yes | FP16 · ~26.3 GB · ~13 tok/s |
| Apple M2 Ultra (192GB) | Yes | FP16 · ~26.3 GB · ~20 tok/s |