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

category chat · Qwen/Qwen3.8-Flash-Next · 19 runs in the model record
answer: 69.4 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 ×2                | 3-bit | 21.9 tok/s     | reported | ttft -ms | vram - GB | ctx 16,384
  Arc Pro B70 32GB ×2                | 4-bit | 24.5 tok/s     | reported | ttft 2,810ms | vram - GB | ctx 49,152
  M3 Ultra 512GB                     | 4-bit | 14 tok/s       | reported | ttft 2,053ms | vram - GB | ctx 2,048
  Radeon AI Pro R9700 32GB ×3        | 3-bit | 31.1 tok/s     | measured | ttft 1,001ms | vram 64 GB | ctx 131,072
  Radeon AI Pro R9700 32GB ×3        | 4-bit | 7.7 tok/s      | reported | ttft 4,017ms | vram - GB | ctx 789
  RTX PRO 6000 Blackwell 96GB ×2     | 8-bit | 125.4 tok/s    | measured | ttft 100ms | vram - GB | ctx 262,144
  Ryzen AI Max 395 128GB             | 4-bit | 54 tok/s       | reported | ttft 6ms | vram 124.1 GB | ctx 262,144