FitLocally by GPUwerk

Mistral · 3.4B

Can you run Ministral 3 3B locally?

Ministral 3 3B is a 3.4B-parameter model. At 4-bit or better it runs on 120 of the 120 hardware configurations listed here; the smallest is the RTX 5060 (8 GB). On a 128 GB DGX Spark it runs at Q4_K_M.

Data as of 2 November 2026

Fastest machines for Ministral 3 3B

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~291 tok/s ModelFit est.Q4_K_M · 4.1 of 31 GB
  2. 2RTX PRO 6000 BlackwellGraphics card · 96 GB~291 tok/s ModelFit est.Q4_K_M · 4.1 of 95 GB
  3. 32x RTX 5090Graphics card · 64 GB~268 tok/s ModelFit est.Q4_K_M · 4.1 of 62 GB
  4. 42x RTX PRO 6000 BlackwellGraphics card · 192 GB~268 tok/s ModelFit est.Q4_K_M · 4.1 of 190 GB
  5. 5RTX 4090Graphics card · 24 GB~192 tok/s ModelFit est.Q4_K_M · 4.1 of 23 GB
  6. 6RTX 5080Graphics card · 16 GB~191 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  7. 7M5 UltraMac · 96–512 GB · 5 variants~187 tok/s ModelFit est.Q4_K_M · 4.1 GB
  8. 82x RTX 4090Graphics card · 48 GB~182 tok/s ModelFit est.Q4_K_M · 4.1 of 46 GB
  9. 9RTX 5070 TiGraphics card · 16 GB~181 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  10. 10RTX 3090 TiGraphics card · 24 GB~178 tok/s ModelFit est.Q4_K_M · 4.1 of 23 GB
  11. 112x RTX 3090 TiGraphics card · 48 GB~169 tok/s ModelFit est.Q4_K_M · 4.1 of 46 GB
  12. 12RTX 3090Graphics card · 24 GB~169 tok/s ModelFit est.Q4_K_M · 4.1 of 23 GB
  13. 132x RTX 3090Graphics card · 48 GB~161 tok/s ModelFit est.Q4_K_M · 4.1 of 46 GB
  14. 14RTX 4080 SUPERGraphics card · 16 GB~151 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  15. 15RTX 4080Graphics card · 16 GB~148 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  16. 16RTX 5070Graphics card · 12 GB~145 tok/s ModelFit est.Q4_K_M · 4.1 of 11.0 GB
  17. 17RTX 4070 Ti SUPERGraphics card · 16 GB~140 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  18. 18RX 7900 XTXGraphics card · 24 GB~140 tok/s ModelFit est.Q4_K_M · 4.1 of 23 GB
  19. 192x RX 7900 XTXGraphics card · 48 GB~135 tok/s ModelFit est.Q4_K_M · 4.1 of 46 GB
  20. 20M3 UltraMac · 96–512 GB · 5 variants~132 tok/s ModelFit est.Q4_K_M · 4.1 GB
  21. 21M5 MaxMac · 48–128 GB · 3 variants~126 tok/s ModelFit est.Q4_K_M · 4.1 GB
  22. 22RX 7900 XTGraphics card · 20 GB~126 tok/s ModelFit est.Q4_K_M · 4.1 of 19.0 GB
  23. 23M2 UltraMac · 64–192 GB · 7 variants~116 tok/s ModelFit est.Q4_K_M · 4.1 GB
  24. 24RTX 4070 TiGraphics card · 12 GB~111 tok/s ModelFit est.Q4_K_M · 4.1 of 11.0 GB
  25. 25RTX 4070 SUPERGraphics card · 12 GB~111 tok/s ModelFit est.Q4_K_M · 4.1 of 11.0 GB
  26. 26RTX 4070Graphics card · 12 GB~111 tok/s ModelFit est.Q4_K_M · 4.1 of 11.0 GB
  27. 27RX 9070 XTGraphics card · 16 GB~110 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  28. 28M4 MaxMac · 48–128 GB · 3 variants~106 tok/s ModelFit est.Q4_K_M · 4.1 GB
  29. 29RTX 5060Graphics card · 8 GB~104 tok/s ModelFit est.Q4_K_M · 4.1 of 7.0 GB
  30. 30RTX 5060 TiGraphics card · 16 GB~104 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  31. 31M1 UltraMac · 64–128 GB · 4 variants~103 tok/s ModelFit est.Q4_K_M · 4.1 GB
  32. 32M5 Max · 36 GB · 32cMac · 36 GB~99 tok/s ModelFit est.Q4_K_M · 4.1 of 24 GB
  33. 33RTX 3070Graphics card · 8 GB~92 tok/s ModelFit est.Q4_K_M · 4.1 of 7.0 GB
  34. 34RTX 3060 TiGraphics card · 8 GB~92 tok/s ModelFit est.Q4_K_M · 4.1 of 7.0 GB
  35. 35M3 MaxMac · 48–128 GB · 3 variants~84 tok/s ModelFit est.Q4_K_M · 4.1 GB
  36. 36RTX 2070Graphics card · 8 GB~83 tok/s ModelFit est.Q4_K_M · 4.1 of 7.0 GB
  37. 37M4 Max · 36 GB · 32cMac · 36 GB~83 tok/s ModelFit est.Q4_K_M · 4.1 of 24 GB
  38. 38M2 MaxMac · 32–96 GB · 5 variants~78 tok/s ModelFit est.Q4_K_M · 4.1 GB
  39. 39RTX 3060Graphics card · 12 GB~76 tok/s ModelFit est.Q4_K_M · 4.1 of 11.0 GB
  40. 40M5 ProMac · 24–64 GB · 6 variants~75 tok/s ModelFit est.Q4_K_M · 4.1 GB
  41. 41M1 MaxMac · 32–64 GB · 4 variants~70 tok/s ModelFit est.Q4_K_M · 4.1 GB
  42. 42RX 9060 XTGraphics card · 16 GB~67 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  43. 43RTX 4060 TiGraphics card · 16 GB~66 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  44. 44Tesla P100Graphics card · 16 GB~65 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  45. 45M3 MaxMac · 36–96 GB · 2 variants~65 tok/s ModelFit est.Q4_K_M · 4.1 GB
  46. 46M4 ProMac · 24–64 GB · 6 variants~63 tok/s ModelFit est.Q4_K_M · 4.1 GB
  47. 47RTX 4060Graphics card · 8 GB~63 tok/s ModelFit est.Q4_K_M · 4.1 of 7.0 GB
  48. 48Tesla T4Graphics card · 16 GB~58 tok/s ModelFit est.Q4_K_M · 4.1 of 15.0 GB
  49. 49Ryzen AI Max+ 395 · 128 GBAI mini PC · 128 GB~49 tok/s ModelFit est.Q4_K_M · 4.1 of 96 GB
  50. 50M2 ProMac · 16–32 GB · 2 variants~46 tok/s ModelFit est.Q4_K_M · 4.1 GB
  51. 51M6Mac · 24–32 GB · 3 variants~45 tok/s ModelFit est.Q4_K_M · 4.1 GB
  52. 52M1 ProMac · 16–32 GB · 2 variants~41 tok/s ModelFit est.Q4_K_M · 4.1 GB
  53. 53M6Mac · 16 GB · 2 variants~40 tok/s ModelFit est.Q4_K_M · 4.1 GB
  54. 54M5Mac · 16–32 GB · 6 variants~38 tok/s ModelFit est.Q4_K_M · 4.1 GB
  55. 55M3 ProMac · 18–36 GB · 2 variants~37 tok/s ModelFit est.Q4_K_M · 4.1 GB
  56. 56M4Mac · 16–32 GB · 3 variants~28 tok/s ModelFit est.Q4_K_M · 4.1 GB
  57. 57M3Mac · 8–24 GB · 3 variants~25 tok/s ModelFit est.Q4_K_M · 4.1 GB
  58. 58M2Mac · 8–24 GB · 3 variants~24 tok/s ModelFit est.Q4_K_M · 4.1 GB
  59. 59M1Mac · 8–16 GB · 2 variants~15 tok/s ModelFit est.Q4_K_M · 4.1 GB
  60. 60Ryzen AI Max+ 395 · 64 GBAI mini PC · 64 GBno estimateQ4_K_M · 4.1 of 48 GB
  61. 61DGX Spark (GB10)AI mini PC · 128 GBno estimateQ4_K_M · 4.1 of 122 GB

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
3.4B
Active per token
not in our sources
Type
not in our sources
Good for
Chat, Vision, Edge

Builds and how to run them

QuantizationLoadMin. RAMCommand
Q4_K_M3.0 GB6 GBollama run ministral-3:3b

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 Ministral 3 3B 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 M3 · 8 GB.

How much memory does Ministral 3 3B need?

About 4.1 GB at the smallest build in our data, including a margin for the runtime and an 8K-token context. Longer contexts need more.

Does Ministral 3 3B run on a DGX Spark?

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