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-3090-24gb-x8.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  rtx-3090-24gb-x8                   | deepseek-ai-deepseek-v4-flash-0731     | reported | 93.3 tok/s     | n=2
  rtx-3090-24gb-x8                   | hell0ks-glm-5-3-flash-uncensored-awq   | reported | 76.7 tok/s     | n=1
  rtx-3090-24gb-x8                   | wtdcode-glm-5-3-flash-awq-w4a16        | reported | 75.2 tok/s     | n=1
  rtx-3090-24gb-x8                   | drowzeys-keys-deepseekv4flash-vision-exp-ablit | reported | 66.9 tok/s     | n=1
  rtx-3090-24gb-x8                   | stepfun-ai-step-3-7-flash-gguf         | reported | 52.7 tok/s     | n=1
  rtx-3090-24gb-x8                   | qwen-qwen3-5-122b-a10b-gptq-int4       | reported | 49.1 tok/s     | n=1
  rtx-3090-24gb-x8                   | unsloth-glm-5-3-flash-gguf             | reported | 35 tok/s       | n=2
  rtx-3090-24gb-x8                   | bartowski-zai-org-glm-5-1-gguf         | reported | 26.1 tok/s     | n=1
  rtx-3090-24gb-x8                   | qwen-qwen3-5-122b-a10b-fp8             | reported | 7.4 tok/s      | n=1
  rtx-3090-24gb-x8                   | bartowski-moonshotai-kimi-k2-5-gguf    | reported | 5.5 tok/s      | n=1

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