Mistral Nemo 12B
mistralai/Mistral-Nemo-Instruct-2407
Parameters
12.2B
Context
128K
Downloads
413.5K
Likes
1.7K
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 33.7 GB RAM 50.5 · Disk 74.5 | 50.8 GB RAM 76.1 · Disk 100.5 | 149.0 GB RAM 223.4 · Disk 139.5 |
| FP88bit | 19.3 GB RAM 32.0 · Disk 56.7 | 35.7 GB RAM 53.6 · Disk 82.7 | 133.3 GB RAM 199.9 · Disk 121.7 |
| INT88bit | 19.3 GB RAM 32.0 · Disk 56.7 | 35.7 GB RAM 53.6 · Disk 82.7 | 133.3 GB RAM 199.9 · Disk 121.7 |
| AWQ4bit | 12.4 GB RAM 32.0 · Disk 48.2 | 28.5 GB RAM 42.7 · Disk 74.2 | 125.7 GB RAM 188.6 · Disk 113.2 |
| GPTQ4bit | 12.5 GB RAM 32.0 · Disk 48.3 | 28.6 GB RAM 42.9 · Disk 74.3 | 125.8 GB RAM 188.8 · Disk 113.3 |
| GGUF4bit | 12.7 GB RAM 32.0 · Disk 48.5 | 28.8 GB RAM 43.2 · Disk 74.5 | 126.0 GB RAM 189.0 · Disk 113.5 |
GPU Recommendations
1× H100 SXM
80 GB VRAM
≈ 568.8 tok/s
Effective production throughput ≈ 398.2 tok/s
score 0.5845
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.6097
8× H200 SXM
1128 GB VRAM
≈ 2607.8 tok/s
Effective production throughput ≈ 1825.5 tok/s
score 0.2197
1× RTX A6000
48 GB VRAM
≈ 130.4 tok/s
Effective production throughput ≈ 91.3 tok/s
score 0.6494
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX A6000
48 GB VRAM
$210
/mo · vast
RTX 5090
32 GB VRAM
$244
/mo · vast
L40S
48 GB VRAM
$341
/mo · vast
RTX PRO 6000 WS
48 GB VRAM
$586
/mo · vast
A100 80GB
80 GB VRAM
$876
/mo · vast