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-3060-ti-8gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  rtx-3060-ti-8gb                    | qwen-qwen3-0-6b                        | measured | 141.7 tok/s    | n=18
  rtx-3060-ti-8gb                    | qwen-qwen3-1-7b-gguf                   | measured | 116 tok/s      | n=12
  rtx-3060-ti-8gb                    | nvidia-nvidia-nemotron-3-nano-4b-bf16  | measured | 112.5 tok/s    | n=27
  rtx-3060-ti-8gb                    | lmstudio-community-openreasoning-nemotron-7b-gguf | measured | 68.1 tok/s     | n=18
  rtx-3060-ti-8gb                    | nvidia-acereason-nemotron-1-1-7b       | measured | 68.1 tok/s     | n=19
  rtx-3060-ti-8gb                    | deepseek-ai-deepseek-r1-distill-qwen-7b | measured | 67.9 tok/s     | n=9
  rtx-3060-ti-8gb                    | nvidia-nvidia-nemotron-nano-9b-v2      | reported | 64.3 tok/s     | n=2
  rtx-3060-ti-8gb                    | qwen-qwen3-5-9b                        | reported | 63.5 tok/s     | n=1
  rtx-3060-ti-8gb                    | qwen-qwen3-8b                          | measured | 53 tok/s       | n=26

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