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

category chat · fraserprice/DeepSeek-V4-Flash-DSpark · 6 runs in the model record
answer: 281.5 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 256GB         | 4-bit | 65.1 tok/s     | reported | ttft 468ms | vram 97.3 GB | ctx 262,144
  RTX PRO 6000 Blackwell 96GB ×2     | 8-bit | 253.6 tok/s    | reported | ttft 42ms | vram - GB | ctx 2,048
  RTX PRO 6000 Blackwell 96GB ×4     | 8-bit | 308.2 tok/s    | measured | ttft 39ms | vram - GB | ctx 2,048