FitLocally by GPUwerk

Granite · 8B

Can you run Granite 4.1 8B Instruct locally?

Granite 4.1 8B Instruct is a 8B-parameter model. At 4-bit or better it runs on 117 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 Granite 4.1 8B Instruct

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~198 tok/s ModelFit est.Q4_K_M · 6.6 of 31 GB
  2. 2RTX PRO 6000 BlackwellGraphics card · 96 GB~198 tok/s ModelFit est.Q4_K_M · 6.6 of 95 GB
  3. 32x RTX 5090Graphics card · 64 GB~187 tok/s ModelFit est.Q4_K_M · 6.6 of 62 GB
  4. 42x RTX PRO 6000 BlackwellGraphics card · 192 GB~187 tok/s ModelFit est.Q4_K_M · 6.6 of 190 GB
  5. 5M5 UltraMac · 96–512 GB · 5 variants~126 tok/s ModelFit est.Q4_K_M · 6.6 GB
  6. 6RTX 4090Graphics card · 24 GB~122 tok/s ModelFit est.Q4_K_M · 6.6 of 23 GB
  7. 7RTX 5080Graphics card · 16 GB~121 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  8. 82x RTX 4090Graphics card · 48 GB~118 tok/s ModelFit est.Q4_K_M · 6.6 of 46 GB
  9. 9RTX 5070 TiGraphics card · 16 GB~115 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  10. 10RTX 3090 TiGraphics card · 24 GB~115 tok/s ModelFit est.Q4_K_M · 6.6 of 23 GB
  11. 112x RTX 3090 TiGraphics card · 48 GB~111 tok/s ModelFit est.Q4_K_M · 6.6 of 46 GB
  12. 12RTX 3090Graphics card · 24 GB~108 tok/s ModelFit est.Q4_K_M · 6.6 of 23 GB
  13. 132x RTX 3090Graphics card · 48 GB~104 tok/s ModelFit est.Q4_K_M · 6.6 of 46 GB
  14. 14RX 7900 XTXGraphics card · 24 GB~96 tok/s ModelFit est.Q4_K_M · 6.6 of 23 GB
  15. 152x RX 7900 XTXGraphics card · 48 GB~94 tok/s ModelFit est.Q4_K_M · 6.6 of 46 GB
  16. 16RTX 4080 SUPERGraphics card · 16 GB~94 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  17. 17RTX 4080Graphics card · 16 GB~91 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  18. 18RTX 5070Graphics card · 12 GB~89 tok/s ModelFit est.Q4_K_M · 6.6 of 11.0 GB
  19. 19M3 UltraMac · 96–512 GB · 5 variants~86 tok/s ModelFit est.Q4_K_M · 6.6 GB
  20. 20RTX 4070 Ti SUPERGraphics card · 16 GB~86 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  21. 21RX 7900 XTGraphics card · 20 GB~84 tok/s ModelFit est.Q4_K_M · 6.6 of 19.0 GB
  22. 22M2 UltraMac · 64–192 GB · 7 variants~77 tok/s ModelFit est.Q4_K_M · 6.6 GB
  23. 23M5 MaxMac · 48–128 GB · 3 variants~76 tok/s ModelFit est.Q4_K_M · 6.6 GB
  24. 24RX 9070 XTGraphics card · 16 GB~73 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  25. 25M1 UltraMac · 64–128 GB · 4 variants~69 tok/s ModelFit est.Q4_K_M · 6.6 GB
  26. 26RTX 4070 TiGraphics card · 12 GB~67 tok/s ModelFit est.Q4_K_M · 6.6 of 11.0 GB
  27. 27RTX 4070 SUPERGraphics card · 12 GB~67 tok/s ModelFit est.Q4_K_M · 6.6 of 11.0 GB
  28. 28RTX 4070Graphics card · 12 GB~67 tok/s ModelFit est.Q4_K_M · 6.6 of 11.0 GB
  29. 29M4 MaxMac · 48–128 GB · 3 variants~64 tok/s ModelFit est.Q4_K_M · 6.6 GB
  30. 30RTX 5060Graphics card · 8 GB~62 tok/s ModelFit est.Q4_K_M · 6.6 of 7.0 GB
  31. 31RTX 5060 TiGraphics card · 16 GB~62 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  32. 32M5 Max · 36 GB · 32cMac · 36 GB~59 tok/s ModelFit est.Q4_K_M · 6.6 of 24 GB
  33. 33RTX 3070Graphics card · 8 GB~55 tok/s ModelFit est.Q4_K_M · 6.6 of 7.0 GB
  34. 34RTX 3060 TiGraphics card · 8 GB~55 tok/s ModelFit est.Q4_K_M · 6.6 of 7.0 GB
  35. 35M3 MaxMac · 48–128 GB · 3 variants~50 tok/s ModelFit est.Q4_K_M · 6.6 GB
  36. 36M4 Max · 36 GB · 32cMac · 36 GB~49 tok/s ModelFit est.Q4_K_M · 6.6 of 24 GB
  37. 37RTX 2070Graphics card · 8 GB~49 tok/s ModelFit est.Q4_K_M · 6.6 of 7.0 GB
  38. 38M2 MaxMac · 32–96 GB · 5 variants~47 tok/s ModelFit est.Q4_K_M · 6.6 GB
  39. 39RTX 3060Graphics card · 12 GB~45 tok/s ModelFit est.Q4_K_M · 6.6 of 11.0 GB
  40. 40M5 ProMac · 24–64 GB · 6 variants~44 tok/s ModelFit est.Q4_K_M · 6.6 GB
  41. 41M1 MaxMac · 32–64 GB · 4 variants~42 tok/s ModelFit est.Q4_K_M · 6.6 GB
  42. 42RX 9060 XTGraphics card · 16 GB~41 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  43. 43RTX 4060 TiGraphics card · 16 GB~39 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  44. 44Tesla P100Graphics card · 16 GB~38 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  45. 45M3 MaxMac · 36–96 GB · 2 variants~38 tok/s ModelFit est.Q4_K_M · 6.6 GB
  46. 46RTX 4060Graphics card · 8 GB~37 tok/s ModelFit est.Q4_K_M · 6.6 of 7.0 GB
  47. 47M4 ProMac · 24–64 GB · 6 variants~37 tok/s ModelFit est.Q4_K_M · 6.6 GB
  48. 48Tesla T4Graphics card · 16 GB~34 tok/s ModelFit est.Q4_K_M · 6.6 of 15.0 GB
  49. 49Ryzen AI Max+ 395 · 128 GBAI mini PC · 128 GB~30 tok/s ModelFit est.Q4_K_M · 6.6 of 96 GB
  50. 50M2 ProMac · 16–32 GB · 2 variants~27 tok/s ModelFit est.Q4_K_M · 6.6 GB
  51. 51M6Mac · 24–32 GB · 3 variants~26 tok/s ModelFit est.Q4_K_M · 6.6 GB
  52. 52M1 ProMac · 16–32 GB · 2 variants~24 tok/s ModelFit est.Q4_K_M · 6.6 GB
  53. 53M6Mac · 16 GB · 2 variants~23 tok/s ModelFit est.Q4_K_M · 6.6 GB
  54. 54M5Mac · 16–32 GB · 6 variants~22 tok/s ModelFit est.Q4_K_M · 6.6 GB
  55. 55M3 ProMac · 18–36 GB · 2 variants~21 tok/s ModelFit est.Q4_K_M · 6.6 GB
  56. 56M4Mac · 16–32 GB · 3 variants~16 tok/s ModelFit est.Q4_K_M · 6.6 GB
  57. 57M3Mac · 16–24 GB · 2 variants~14 tok/s ModelFit est.Q4_K_M · 6.6 GB
  58. 58M2Mac · 16–24 GB · 2 variants~14 tok/s ModelFit est.Q4_K_M · 6.6 GB
  59. 59M1 · 16 GBMac · 16 GB~8.4 tok/s slow ModelFit est.Q4_K_M · 6.6 of 10.7 GB
  60. 60Ryzen AI Max+ 395 · 64 GBAI mini PC · 64 GBno estimateQ4_K_M · 6.6 of 48 GB
  61. 61DGX Spark (GB10)AI mini PC · 128 GBno estimateQ4_K_M · 6.6 of 122 GB
Too big for 3 machines

M3 · 8 GB, M2 · 8 GB, M1 · 8 GB. Granite 4.1 8B Instruct needs about 6.6 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
8B
Active per token
not in our sources
Type
not in our sources
Good for
Enterprise assistant, tool calling, instruction following

Builds and how to run them

QuantizationLoadMin. RAMCommand
Q4_K_M5.5 GB8 GBollama run granite4.1:8b

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 Granite 4.1 8B Instruct 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 Pro · 16 GB.

How much memory does Granite 4.1 8B Instruct need?

About 6.6 GB at the smallest build in our data, including a margin for the runtime and an 8K-token context. Longer contexts need more.

Does Granite 4.1 8B Instruct run on a DGX Spark?

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