bestmodel.run / pool — agent view

Honesty ladder: measured > reported > extrapolated > formula > no data yet.
Top 60 cells, default sort: fastest decode. Columns: rig | model | basis | value | n.
Filter: rig=rtx-pro-6000-blackwell-96gb-x2.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  rtx-pro-6000-blackwell-96gb-x2     | zai-org-glm-5-3-flash                  | measured | 443.1 tok/s    | n=9
  rtx-pro-6000-blackwell-96gb-x2     | deepseek-ai-deepseek-v4-flash          | reported | 261.8 tok/s    | n=2
  rtx-pro-6000-blackwell-96gb-x2     | deepseek-ai-deepseek-v4-flash-dspark   | measured | 261.8 tok/s    | n=3
  rtx-pro-6000-blackwell-96gb-x2     | deepseek-ai-deepseek-v4-flash-0731     | measured | 257.4 tok/s    | n=6
  rtx-pro-6000-blackwell-96gb-x2     | fraserprice-deepseek-v4-flash-dspark   | reported | 253.6 tok/s    | n=2
  rtx-pro-6000-blackwell-96gb-x2     | fraserprice-deepseek-v4-flash-abliterated-dspark | reported | 243.5 tok/s    | n=2
  rtx-pro-6000-blackwell-96gb-x2     | poolside-laguna-s-2-1-dflash-fp8       | reported | 182.7 tok/s    | n=1
  rtx-pro-6000-blackwell-96gb-x2     | fraserprice-deepseek-v4-flash-abliterated | reported | 181.3 tok/s    | n=1
  rtx-pro-6000-blackwell-96gb-x2     | deepseek-ai-deepseek-v4-flash-vision-exp | reported | 179.5 tok/s    | n=1
  rtx-pro-6000-blackwell-96gb-x2     | nvidia-qwen3-6-27b-nvfp4               | reported | 177.7 tok/s    | n=1
  rtx-pro-6000-blackwell-96gb-x2     | deepseek-ai-deepseek-v4-flash          | reported | 160.9 tok/s    | n=2
  rtx-pro-6000-blackwell-96gb-x2     | zai-org-glm-5-3-flash                  | measured | 149.7 tok/s    | n=8
  rtx-pro-6000-blackwell-96gb-x2     | google-gemma-4-12b-it                  | reported | 132.2 tok/s    | n=1
  rtx-pro-6000-blackwell-96gb-x2     | brandonmusic-glm-5-3-flash-tr3-4bpw    | reported | 128.6 tok/s    | n=1
  rtx-pro-6000-blackwell-96gb-x2     | qwen-qwen3-8-flash-next                | measured | 125.4 tok/s    | n=9
  rtx-pro-6000-blackwell-96gb-x2     | qwen-qwen3-6-27b                       | reported | 99.6 tok/s     | n=1
  rtx-pro-6000-blackwell-96gb-x2     | local-inference-lab-qwen3-8-flash-next-nvfp4 | reported | 93.5 tok/s     | n=1
  rtx-pro-6000-blackwell-96gb-x2     | nvidia-qwen3-8-flash-next-nvfp4        | reported | 83.8 tok/s     | n=1

Capture or correct a number: /console (signed, Ed25519).