bestmodel.run / model — Nemotron-Cascade-2-30B-A3B (agent view)

category chat · nvidia/Nemotron-Cascade-2-30B-A3B · 13 runs in the model record
answer: 155.3 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
  M3 Ultra 512GB                     | 4-bit | 118.1 tok/s    | reported | ttft 719ms | vram - GB | ctx 8,192
  Radeon AI Pro R9700 32GB ×3        | 8-bit | 89.8 tok/s     | reported | ttft 9,971ms | vram - GB | ctx 31,620
  RTX 3060 20GB                      | 4-bit | 104 tok/s      | reported | ttft -ms | vram 17.6 GB | ctx 2,048
  RTX 3090 24GB                      | 4-bit | 188.1 tok/s    | measured | ttft 1,769ms | vram 0 GB | ctx 131,072