Home / Models / Mixtral 8x22B Instruct
Mixtral 8x22B Instruct
141B parameters · 39B active · native context 65,536 · mixtral
Estimated VRAM requirement
Weight memory uses parameter count × bits per weight, plus a quantization overhead factor. KV cache uses standard GQA formulas. Values are calculated estimates, not measured allocations.
| Quantization | Estimated weights | 2K total | 4K total | 8K total | 16K total | 32K total | 64K total |
|---|---|---|---|---|---|---|---|
| FP16 | 267.9 GB | 290.5 GB | 290.9 GB | 291.8 GB | 293.6 GB | 297.1 GB | 304.1 GB |
| Q8 | 150.7 GB | 163.9 GB | 164.4 GB | 165.2 GB | 167.0 GB | 170.5 GB | 177.5 GB |
| Q6 | 117.4 GB | 127.9 GB | 128.4 GB | 129.3 GB | 131.0 GB | 134.5 GB | 141.5 GB |
| Q5 | 99.3 GB | 108.4 GB | 108.9 GB | 109.8 GB | 111.5 GB | 115.0 GB | 122.0 GB |
| Q4 | 82.7 GB | 90.5 GB | 91.0 GB | 91.8 GB | 93.6 GB | 97.1 GB | 104.1 GB |
| Q3 | 66.1 GB | 72.5 GB | 73.0 GB | 73.9 GB | 75.6 GB | 79.1 GB | 86.1 GB |
Find a GPU for this model
Run at the quantization and context selected below. Defaults are Q4 and 8K when you do not change the controls. Results are full-VRAM fits only — not a speed ranking.
- Run at
- Q4 · 8K context
- Required VRAM
- —
No published GPU calculates as a full VRAM fit for Q4 at 8K.
Compatible GPUs
Status is a calculated memory fit against usable VRAM (advertised × 0.9). Open a GPU to see product SKUs and Amazon listings.
Sources
Model specifications: official repository mistralai/Mixtral-8x22B-Instruct-v0.1 · revision main. Compatibility: RigForAI calculator v1.0.0 (estimated VRAM). GPU products: Icecat. Amazon is a separate affiliate match and is not used as a spec source.