bestmodel.run / model — Gemma-4-26B-A4B-NVFP4 (agent view)

category chat · nvidia/Gemma-4-26B-A4B-NVFP4 · 5 runs in the model record
answer: 442 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
  GB10 Grace Blackwell 128GB         | 4-bit | 130.5 tok/s    | reported | ttft 200ms | vram - GB | ctx 2,048
  M1 Max 64GB                        | 4-bit | 52.7 tok/s     | reported | ttft -ms | vram - GB | ctx 2,048
  RTX PRO 6000 Blackwell Server Edition 96GB | 4-bit | 128.3 tok/s    | reported | ttft -ms | vram 87.5 GB | ctx 65,536
  RTX PRO 6000 Blackwell Workstation Edition 96GB | 4-bit | 36.9 tok/s     | reported | ttft 60ms | vram - GB | ctx 131,072