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-2080-8gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  rtx-2080-8gb                       | qwen-qwen2-5-coder-7b-instruct         | reported | 93 tok/s       | n=1
  rtx-2080-8gb                       | deepseek-ai-deepseek-r1-distill-qwen-7b | reported | 80 tok/s       | n=1
  rtx-2080-8gb                       | meta-llama-llama-3-1-8b-instruct       | reported | 76 tok/s       | n=1
  rtx-2080-8gb                       | deepseek-ai-deepseek-coder-v2-lite-instruct | reported | 39 tok/s       | n=1
  rtx-2080-8gb                       | google-gemma-3n-e4b-it                 | reported | 34 tok/s       | n=1
  rtx-2080-8gb                       | google-gemma-4-e4b-it                  | reported | 29 tok/s       | n=2
  rtx-2080-8gb                       | google-gemma-3-12b-it                  | reported | 21 tok/s       | n=2
  rtx-2080-8gb                       | qwen-qwen3-30b-a3b                     | reported | 19 tok/s       | n=1
  rtx-2080-8gb                       | openai-gpt-oss-20b                     | reported | 12 tok/s       | n=1
  rtx-2080-8gb                       | qwen-qwen2-5-coder-14b-instruct        | reported | 11 tok/s       | n=2
  rtx-2080-8gb                       | qwen-qwen2-5-14b-instruct              | reported | 10 tok/s       | n=2

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