bestmodel.run / model — Qwen3.8-Flash-Next-GGUF (agent view)

category chat · unsloth/Qwen3.8-Flash-Next-GGUF · 17 runs in the model record
answer: 44 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
  CMP 170HX 64GB                     | 1-bit | 44.6 tok/s     | measured | ttft 321ms | vram 42.1 GB | ctx 65,536
  GB10 Grace Blackwell 128GB         | 4-bit | 27.3 tok/s     | reported | ttft -ms | vram - GB | ctx 131,072
  M5 Max 128GB                       | 3-bit | 41.5 tok/s     | reported | ttft 289ms | vram - GB | ctx 2,048
  M5 Max 128GB                       | 4-bit | 39.4 tok/s     | reported | ttft 304ms | vram - GB | ctx 2,048
  Radeon AI Pro R9700 32GB           | 4-bit | 25.6 tok/s     | reported | ttft 4,747ms | vram 32.9 GB | ctx 32,768
  RTX 3090 24GB ×2                   | 3-bit | 41.6 tok/s     | reported | ttft 6,505ms | vram 43.6 GB | ctx 217,088
  RTX 5090 32GB ×2                   | 4-bit | 142 tok/s      | reported | ttft 515ms | vram 59.9 GB | ctx 131,072
  Ryzen AI Max 395 128GB             | 4-bit | 74.9 tok/s     | measured | ttft 1,356ms | vram 68.3 GB | ctx 32,768