NVIDIA H200 SXM 141GB meets the minimum VRAM requirement for Q4 inference of mistralai/Ministral-3-14B-Instruct-2512. Review the quantization breakdown below to see how higher precision settings impact VRAM and throughput.
NVIDIA H200 SXM 141GB can run mistralai/Ministral-3-14B-Instruct-2512 with Q4 quantization. At approximately 500 tokens/second, you can expect Excellent speed - conversational response times under 1 second.
You have 133GB headroom, which is sufficient for system overhead and smooth operation.
| Quantization | VRAM needed | VRAM available | Estimated speed | Verdict |
|---|---|---|---|---|
| Q4 | 8GB | 141GB | 499.74 tok/s | ✅ Fits comfortably |
| Q8 | 16GB | 141GB | 359.56 tok/s | ✅ Fits comfortably |
| FP16 | 32GB | 141GB | 184.34 tok/s | ✅ Fits comfortably |