Mixtral-8x7B-Instruct-v0.1
mistralai/Mixtral-8x7B-Instruct-v0.1
First-generation MoE (8x7B), efficient sparse inference.
Parameters
46.7B
Context
32K
Downloads
2M
Likes
3.4K
Architecture
mixtral
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 115.0 GB RAM 172.5 · Disk 174.7 | 135.7 GB RAM 203.6 · Disk 200.7 | 237.6 GB RAM 356.5 · Disk 239.7 |
| FP88bit | 60.0 GB RAM 90.0 · Disk 106.9 | 78.2 GB RAM 117.3 · Disk 132.9 | 177.6 GB RAM 266.4 · Disk 171.9 |
| INT88bit | 60.0 GB RAM 90.0 · Disk 106.9 | 78.2 GB RAM 117.3 · Disk 132.9 | 177.6 GB RAM 266.4 · Disk 171.9 |
| AWQ4bit | 33.5 GB RAM 50.2 · Disk 74.2 | 50.5 GB RAM 75.8 · Disk 100.2 | 148.7 GB RAM 223.1 · Disk 139.2 |
| GPTQ4bit | 33.8 GB RAM 50.8 · Disk 74.6 | 50.9 GB RAM 76.4 · Disk 100.6 | 149.1 GB RAM 223.7 · Disk 139.6 |
| GGUF4bit | 34.5 GB RAM 51.8 · Disk 75.5 | 51.6 GB RAM 77.4 · Disk 101.5 | 149.9 GB RAM 224.8 · Disk 140.5 |
GPU Recommendations
1× H100 SXM
80 GB VRAM
≈ 148.5 tok/s
Effective production throughput ≈ 103.9 tok/s
score 0.6531
4× Tesla V100
64 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.584
8× H200 SXM
1128 GB VRAM
≈ 680.9 tok/s
Effective production throughput ≈ 476.6 tok/s
score 0.2197
1× A100 SXM4
80 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.6511
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
A100 SXM4
80 GB VRAM
$354
/mo · vast
A100 PCIE
80 GB VRAM
$446
/mo · vast
A100 80GB
80 GB VRAM
$876
/mo · vast
H100 SXM
80 GB VRAM
$1,404
/mo · vast
H200 NVL
141 GB VRAM
$1,613
/mo · vast