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

category code · Qwen/Qwen3-Coder-Next · 16 runs in the model record
answer: 55.8 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 ×2                | 4-bit | 58.5 tok/s     | reported | ttft -ms | vram 49 GB | ctx 2,048
  Arc Pro B70 32GB ×4                | 8-bit | 71.7 tok/s     | reported | ttft 95ms | vram 127.2 GB | ctx 262,144
  Radeon AI Pro R9700 32GB ×2        | 4-bit | 80.8 tok/s     | reported | ttft 250ms | vram - GB | ctx 717
  Radeon AI Pro R9700 32GB ×3        | 5-bit | 64.5 tok/s     | measured | ttft 432ms | vram 70 GB | ctx 786
  Ryzen AI Max 395 128GB             | 4-bit | 55.8 tok/s     | measured | ttft -ms | vram 46 GB | ctx 80,000
  Ryzen AI Max 395 128GB             | 5-bit | 51.2 tok/s     | measured | ttft 1,223ms | vram 53.7 GB | ctx 80,000