bestmodel.run / model — Qwen3.6-27B-GGUF (agent view)

category chat · unsloth/Qwen3.6-27B-GGUF · 50 runs in the model record
answer: 24.8 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
  Arc Pro B70 32GB                   | 4-bit | 26 tok/s       | measured | ttft -ms | vram - GB | ctx 2,048
  M4 Pro 48GB                        | 4-bit | 12 tok/s       | reported | ttft -ms | vram - GB | ctx 640
  M5 Pro 64GB                        | 4-bit | 11.4 tok/s     | reported | ttft 911ms | vram - GB | ctx 2,048
  Radeon AI Pro R9700 32GB           | 4-bit | 30.2 tok/s     | reported | ttft -ms | vram - GB | ctx 65,536
  Radeon RX 7900 XTX 24GB            | 3-bit | 33.3 tok/s     | reported | ttft -ms | vram 13.5 GB | ctx 640
  Radeon RX 7900 XTX 24GB            | 4-bit | 28.5 tok/s     | reported | ttft -ms | vram 14.8 GB | ctx 640
  RTX 3060 12GB ×2                   | 4-bit | 18.4 tok/s     | reported | ttft 490ms | vram - GB | ctx 32,768
  RTX 3090 24GB                      | 4-bit | 150.1 tok/s    | reported | ttft -ms | vram 19.5 GB | ctx 2,048
  RTX 3090 24GB ×2                   | 4-bit | 37.8 tok/s     | measured | ttft 1,416ms | vram 12.2 GB | ctx 65,536
  RTX 3090 24GB ×2                   | 6-bit | 22.2 tok/s     | reported | ttft 1,532ms | vram 17.9 GB | ctx 32,768
  RTX 3090 24GB ×2                   | 8-bit | 23 tok/s       | reported | ttft 1,659ms | vram 22.1 GB | ctx 32,768
  RTX 3090 Ti 24GB                   | 4-bit | 35.7 tok/s     | measured | ttft 374ms | vram 21 GB | ctx 73,728
  RTX 4070 12GB                      | 2-bit | 21.1 tok/s     | reported | ttft 1,391ms | vram 11.9 GB | ctx 98,304
  RTX 4090 24GB                      | 4-bit | 41.8 tok/s     | reported | ttft -ms | vram - GB | ctx 32,768
  RTX 5060 Ti 16GB ×2                | 4-bit | 22.3 tok/s     | reported | ttft 177ms | vram - GB | ctx 8,192
  RTX 5090 32GB                      | 3-bit | 73.1 tok/s     | measured | ttft 466ms | vram 18.4 GB | ctx 131,072
  RTX 5090 32GB                      | 4-bit | 66.8 tok/s     | measured | ttft 456ms | vram 19.5 GB | ctx 262,144
  RTX 5090 32GB                      | 5-bit | 60.4 tok/s     | measured | ttft 476ms | vram 21.7 GB | ctx 131,072
  RTX 5090 32GB                      | 6-bit | 50.9 tok/s     | measured | ttft 523ms | vram 24.6 GB | ctx 65,536
  RTX A5000 24GB                     | 4-bit | 32.3 tok/s     | reported | ttft -ms | vram 16.2 GB | ctx 4,096
  RX 7900 XTX 24GB                   | 3-bit | 31.7 tok/s     | reported | ttft -ms | vram 11.6 GB | ctx 512
  Tesla P100 32GB ×2                 | 16-bit | 17.2 tok/s     | measured | ttft 10,079ms | vram - GB | ctx 2,048