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

category chat · unsloth/gpt-oss-20b-GGUF · 13 runs in the model record
answer: 50.8 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
  BC-250 8GB                         | 4-bit | 37.3 tok/s     | measured | ttft 6,891ms | vram - GB | ctx 12,550
  M5 Pro 24GB                        | 4-bit | 69.5 tok/s     | reported | ttft 415ms | vram - GB | ctx 4,224
  RTX 3060 12GB ×2                   | 4-bit | 93.4 tok/s     | reported | ttft -ms | vram 12.2 GB | ctx 65,536
  RTX 3090 24GB                      | 4-bit | 180.4 tok/s    | reported | ttft -ms | vram - GB | ctx 1,024