bestmodel.run / model — gpt-oss-20b (agent view)

category chat · openai/gpt-oss-20b · 41 runs in the model record
answer: 106.9 tok/s (model median)

Honesty ladder: measured > reported > extrapolated > formula > no data yet.
Empty fields stay empty — never filled with an estimate.

  rig | quant | median result | basis | ttft | peak vram | max context
  Arc Pro B60 24GB                   | 8-bit | 40.7 tok/s     | reported | ttft 459ms | vram - GB | ctx 4,096
  Arc Pro B60 24GB ×2                | 8-bit | 40.1 tok/s     | reported | ttft 460ms | vram - GB | ctx 4,096
  GB10 Grace Blackwell 121GB         | 16-bit | 49.1 tok/s     | reported | ttft -ms | vram - GB | ctx 128,000
  GB10 Grace Blackwell 128GB         | 4-bit | 49.2 tok/s     | reported | ttft 74ms | vram - GB | ctx 32,768
  M1 Max 64GB                        | 4-bit | 62.4 tok/s     | reported | ttft 172ms | vram - GB | ctx 2,048
  M1 Max 64GB                        | 8-bit | 83.6 tok/s     | reported | ttft -ms | vram - GB | ctx 640
  M5 Max 128GB                       | 4-bit | 137 tok/s      | reported | ttft -ms | vram 12.4 GB | ctx 2,048
  Radeon 890M 32GB                   | 4-bit | 23.9 tok/s     | reported | ttft 1,210ms | vram - GB | ctx 2,048
  Radeon AI Pro R9700 32GB ×2        | 4-bit | 48.3 tok/s     | reported | ttft 89ms | vram - GB | ctx 32,768
  Radeon AI Pro R9700 32GB ×3        | 8-bit | 152.9 tok/s    | measured | ttft 246ms | vram 17 GB | ctx 131,072
  RTX 2080 8GB                       | 4-bit | 12 tok/s       | reported | ttft -ms | vram 12 GB | ctx 2,048
  RTX 3060 12GB                      | 4-bit | 74 tok/s       | reported | ttft -ms | vram 12 GB | ctx 2,048
  RTX 3080 12GB                      | 4-bit | 21 tok/s       | reported | ttft -ms | vram 8.2 GB | ctx 2,048
  RTX 3090 24GB                      | 4-bit | 162.4 tok/s    | measured | ttft 1,790ms | vram 0 GB | ctx 131,072
  RTX 5080 16GB                      | 8-bit | 222.4 tok/s    | reported | ttft 48ms | vram - GB | ctx 4,096
  RTX 5090 32GB                      | 4-bit | 307.2 tok/s    | reported | ttft 83ms | vram - GB | ctx 2,048
  RTX3060+RTX2080-pooled-20GB 20GB   | 4-bit | 77 tok/s       | reported | ttft -ms | vram 12 GB | ctx 2,048
  RX 6700 XT 12GB                    | 8-bit | 22.4 tok/s     | measured | ttft -ms | vram - GB | ctx 4,736
  RX 7900 XTX 24GB                   | 8-bit | 187.7 tok/s    | reported | ttft 231ms | vram 11.1 GB | ctx 8,192
  Ryzen AI Max 385 24GB              | 4-bit | 23.2 tok/s     | reported | ttft 1,210ms | vram - GB | ctx 2,048
  Ryzen AI Max 395 64GB              | 4-bit | 75.2 tok/s     | reported | ttft 931ms | vram 12.2 GB | ctx 8,192