bestmodel.run / model — DeepSeek-V4-Flash-0731-NVFP4 (agent view)

category chat · auroter/DeepSeek-V4-Flash-0731-NVFP4 · 1 runs in the model record
answer: 118.2 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 PRO 6000 Blackwell 96GB ×4     | 4-bit | 118.2 tok/s    | reported | ttft 33ms | vram 354.1 GB | ctx 262,144