bestmodel.run / model — Qwen3.6-27B-Text-NVFP4-MTP (agent view)

category chat · sakamakismile/Qwen3.6-27B-Text-NVFP4-MTP · 6 runs in the model record
answer: 95.7 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 5090 32GB                      | 4-bit | 121.9 tok/s    | reported | ttft 131ms | vram 30 GB | ctx 131,072
  RTX 5090 32GB ×2                   | 4-bit | 85.3 tok/s     | reported | ttft 2,752ms | vram 59.3 GB | ctx 262,144
  RTX PRO 6000 Blackwell 96GB        | 4-bit | 117.4 tok/s    | reported | ttft 75ms | vram - GB | ctx 262,144