bestmodel.run / model — Qwen3-1.7B (agent view)

category chat · Qwen/Qwen3-1.7B · 2 runs in the model record
answer: 180.4 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
  M3 Ultra 512GB                     | 4-bit | 213 tok/s      | reported | ttft 24ms | vram - GB | ctx 8,192
  RTX 3060 12GB                      | 8-bit | 147.7 tok/s    | reported | ttft 548ms | vram 4 GB | ctx 4,608