Mistral AI · 3B · 密集（Dense）
Current-gen tiny Ministral — edge chat with 256K context 看看你的 GPU 或 Mac 跑不跑得動 Ministral 3 3B——最低 1.7 GB，建議 2.8 GB。
2025-12256K context
| 量化 | 位元 | VRAM | 品質 | 狀態 |
|---|---|---|---|---|
| Q2_K | 2 | 1.5 GB | low | — |
| Q3_K_M | 3 | 1.8 GB | moderate | — |
| Q4_K_M | 4 | 2 GB | good | — |
| Q5_K_M | 5 | 2.4 GB | good | — |
| Q6_K | 6 | 2.8 GB | excellent | — |
| Q8_0 | 8 | 3.6 GB | excellent | — |
| F16 | 16 | 6.6 GB | lossless | — |
Can I run Ministral 3 3B locally?
- Can I run Ministral 3 3B locally?
- Ministral 3 3B needs about 1.7 GB of memory at a minimum and 2.8 GB recommended. Open this page to grade it against your GPU or Mac, then run it with runai, Ollama or LM Studio.
- How much VRAM does Ministral 3 3B need?
- At Q4_K_M, Ministral 3 3B uses about 2 GB of VRAM. Higher quants need more memory; lower quants fit tighter cards with a quality tradeoff.