bestmodel.run / model — Laguna-S-2.1-NVFP4 (agent view)

category chat · poolside/Laguna-S-2.1-NVFP4 · 8 runs in the model record
answer: 101.4 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 121GB         | 4-bit | 20.6 tok/s     | reported | ttft 426ms | vram - GB | ctx 128,000
  GB10 Grace Blackwell 128GB         | 4-bit | 73.8 tok/s     | measured | ttft 328ms | vram 109 GB | ctx 262,144
  RTX 5090 32GB ×4                   | 4-bit | 157.1 tok/s    | reported | ttft -ms | vram - GB | ctx 262,144
  RTX PRO 6000 Blackwell 96GB        | 4-bit | 149.6 tok/s    | reported | ttft 57ms | vram - GB | ctx 2,048