bestmodel.run / model — GLM-5.3-Flash (agent view)

category chat · zai-org/GLM-5.3-Flash · 37 runs in the model record
answer: 122.7 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 | 23.5 tok/s     | reported | ttft -ms | vram 99.1 GB | ctx 262,144
  GB10 Grace Blackwell 256GB         | 3-bit | 32.6 tok/s     | measured | ttft 419ms | vram 105 GB | ctx 1,000,000
  GB10 Grace Blackwell 512GB         | 3-bit | 79.7 tok/s     | measured | ttft 664ms | vram - GB | ctx 300,000
  M3 Ultra 512GB                     | 4-bit | 32.9 tok/s     | reported | ttft 1,667ms | vram - GB | ctx 2,048
  Radeon AI Pro R9700 32GB ×3        | 2-bit | 25.7 tok/s     | measured | ttft 3,930ms | vram 87.4 GB | ctx 131,072
  RTX PRO 6000 Blackwell 96GB ×2     | 3-bit | 149.7 tok/s    | measured | ttft 191ms | vram 186.9 GB | ctx 700,000
  RTX PRO 6000 Blackwell 96GB ×2     | 4-bit | 443.1 tok/s    | measured | ttft 125ms | vram 186.9 GB | ctx 8,192
  RTX PRO 6000 Blackwell 96GB ×4     | 8-bit | 146.8 tok/s    | reported | ttft 90ms | vram - GB | ctx 2,048