bestmodel.run / model — Qwen3.5-4B-GGUF (agent view)

category chat · unsloth/Qwen3.5-4B-GGUF · 2 runs in the model record
answer: 68 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
  RTX 3060 12GB                      | 4-bit | 78.6 tok/s     | reported | ttft -ms | vram 3.5 GB | ctx 32,768
  RTX 5070 12GB                      | 16-bit | 68 tok/s       | reported | ttft 77ms | vram - GB | ctx 32,768