FitLocally by GPUwerk

Mistral · 22B

Can you run Mistral Small 22B locally?

Mistral Small 22B is a dense model with 22B parameters. At 4-bit or better it runs on 74 of the 120 hardware configurations listed here; the smallest is the RX 7900 XT (20 GB). On a 128 GB DGX Spark it runs at Q4_K_M, at roughly 8.8 tokens per second (our est.).

Data as of 2 November 2026

Fastest machines for Mistral Small 22B

Every machine that runs it, fastest first. Filter by type; your machine is marked. Compare machines running it side by side

  1. 1RTX 5090Graphics card · 32 GB~105 tok/s ModelFit est.Q4_K_M · 18.4 of 31 GB
  2. 2RTX PRO 6000 BlackwellGraphics card · 96 GB~105 tok/s ModelFit est.Q4_K_M · 18.4 of 95 GB
  3. 32x RTX 5090Graphics card · 64 GB~102 tok/s ModelFit est.Q4_K_M · 18.4 of 62 GB
  4. 42x RTX PRO 6000 BlackwellGraphics card · 192 GB~102 tok/s ModelFit est.Q4_K_M · 18.4 of 190 GB
  5. 5M5 UltraMac · 96–512 GB · 5 variants~65 tok/s ModelFit est.Q4_K_M · 18.4 GB
  6. 6RTX 4090Graphics card · 24 GB~61 tok/s ModelFit est.Q4_K_M · 18.4 of 23 GB
  7. 72x RTX 4090Graphics card · 48 GB~60 tok/s ModelFit est.Q4_K_M · 18.4 of 46 GB
  8. 8RTX 3090 TiGraphics card · 24 GB~57 tok/s ModelFit est.Q4_K_M · 18.4 of 23 GB
  9. 92x RTX 3090 TiGraphics card · 48 GB~56 tok/s ModelFit est.Q4_K_M · 18.4 of 46 GB
  10. 10RTX 3090Graphics card · 24 GB~54 tok/s ModelFit est.Q4_K_M · 18.4 of 23 GB
  11. 112x RTX 3090Graphics card · 48 GB~53 tok/s ModelFit est.Q4_K_M · 18.4 of 46 GB
  12. 12RX 7900 XTXGraphics card · 24 GB~51 tok/s ModelFit est.Q4_K_M · 18.4 of 23 GB
  13. 132x RX 7900 XTXGraphics card · 48 GB~50 tok/s ModelFit est.Q4_K_M · 18.4 of 46 GB
  14. 14M3 UltraMac · 96–512 GB · 5 variants~44 tok/s ModelFit est.Q4_K_M · 18.4 GB
  15. 15RX 7900 XTGraphics card · 20 GB~44 tok/s ModelFit est.Q4_K_M · 18.4 of 19.0 GB
  16. 16M2 UltraMac · 64–192 GB · 7 variants~40 tok/s ModelFit est.Q4_K_M · 18.4 GB
  17. 17M5 MaxMac · 48–128 GB · 3 variants~36 tok/s ModelFit est.Q4_K_M · 18.4 GB
  18. 18M1 UltraMac · 64–128 GB · 4 variants~36 tok/s ModelFit est.Q4_K_M · 18.4 GB
  19. 19M4 MaxMac · 48–128 GB · 3 variants~30 tok/s ModelFit est.Q4_K_M · 18.4 GB
  20. 20M5 Max · 36 GB · 32cMac · 36 GB~27 tok/s ModelFit est.Q4_K_M · 18.4 of 24 GB
  21. 21M3 MaxMac · 48–128 GB · 3 variants~23 tok/s ModelFit est.Q4_K_M · 18.4 GB
  22. 22M4 Max · 36 GB · 32cMac · 36 GB~23 tok/s ModelFit est.Q4_K_M · 18.4 of 24 GB
  23. 23M2 MaxMac · 32–96 GB · 5 variants~22 tok/s ModelFit est.Q4_K_M · 18.4 GB
  24. 24M1 MaxMac · 32–64 GB · 4 variants~20 tok/s ModelFit est.Q4_K_M · 18.4 GB
  25. 25M5 ProMac · 48–64 GB · 4 variants~20 tok/s ModelFit est.Q4_K_M · 18.4 GB
  26. 26M3 MaxMac · 36–96 GB · 2 variants~18 tok/s ModelFit est.Q4_K_M · 18.4 GB
  27. 27M4 ProMac · 48–64 GB · 4 variants~17 tok/s ModelFit est.Q4_K_M · 18.4 GB
  28. 28Ryzen AI Max+ 395 · 128 GBAI mini PC · 128 GB~14 tok/s ModelFit est.Q4_K_M · 18.4 of 96 GB
  29. 29M2 Pro · 32 GBMac · 32 GB~12 tok/s ModelFit est.Q4_K_M · 18.4 of 21 GB
  30. 30M6 · 32 GB · 12cMac · 32 GB~12 tok/s ModelFit est.Q4_K_M · 18.4 of 21 GB
  31. 31M1 Pro · 32 GBMac · 32 GB~11 tok/s ModelFit est.Q4_K_M · 18.4 of 21 GB
  32. 32M5Mac · 32 GB · 2 variants~10 tok/s ModelFit est.Q4_K_M · 18.4 GB
  33. 33M3 Pro · 36 GBMac · 36 GB~9.7 tok/s slow ModelFit est.Q4_K_M · 18.4 of 24 GB
  34. 34DGX Spark (GB10)AI mini PC · 128 GB~8.8 tok/s slow our est.Q4_K_M · 18.4 of 122 GB
  35. 35Ryzen AI Max+ 395 · 64 GBAI mini PC · 64 GB~8.3 tok/s slow our est.Q4_K_M · 18.4 of 48 GB
  36. 36M4 · 32 GBMac · 32 GB~7.4 tok/s slow ModelFit est.Q4_K_M · 18.4 of 21 GB
Too big for 46 machines

M5 Pro · 24 GB · 20c, M5 Pro · 24 GB · 16c, M4 Pro · 24 GB · 20c, M4 Pro · 24 GB · 16c, M6 · 24 GB · 12c, M6 · 24 GB · 10c, M5 · 24 GB · 10c, M5 · 24 GB · 8c, M4 · 24 GB, M3 · 24 GB, M2 · 24 GB, RTX 5080, RTX 5070 Ti, RTX 5060 Ti, RTX 4080 SUPER, RTX 4080, RTX 4070 Ti SUPER, RTX 4060 Ti, Tesla T4, Tesla P100, RX 9070 XT, RX 9060 XT, M3 Pro · 18 GB, RTX 5070, RTX 4070 Ti, RTX 4070 SUPER, RTX 4070, RTX 3060, M2 Pro · 16 GB, M1 Pro · 16 GB, M6 · 16 GB · 12c, M6 · 16 GB · 10c, M5 · 16 GB · 10c, M5 · 16 GB · 8c, M4 · 16 GB, M3 · 16 GB, M2 · 16 GB, M1 · 16 GB, RTX 5060, RTX 4060, RTX 3070, RTX 3060 Ti, RTX 2070, M3 · 8 GB, M2 · 8 GB, M1 · 8 GB. Mistral Small 22B needs about 18.4 GB at its smallest build.

Speeds are single-user decode estimates, not measurements. "ModelFit est." comes from ModelFit's model fitted on published runs (CC BY 4.0); "our est." is memory bandwidth divided by bytes read per token, times 0.55. Memory is what the best build needs, against what the machine can give a model. Method

About the model

Parameters
22B
Active per token
22B
Type
Dense
Good for
Coding, Quality

Builds and how to run them

QuantizationLoadMin. RAMCommand
Q4_K_M17.0 GB32 GBollama run mistral-small:22b

Load, minimum RAM and Ollama commands: ModelFit, CC BY 4.0. Load is memory with the model loaded, before long contexts.

Questions

Can I run Mistral Small 22B on a Mac?

Yes on a Mac with enough memory: the smallest Apple configuration listed here that runs it at 4-bit or better is the M2 Max · 32 GB · 38c.

How much memory does Mistral Small 22B need?

About 18.4 GB at the smallest build in our data, including a margin for the runtime and an 8K-token context. Longer contexts need more.

Does Mistral Small 22B run on a DGX Spark?

Yes, at Q4_K_M. The Spark has 128 GB of unified memory, about 121.6 GB usable.