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=m4-pro-48gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  m4-pro-48gb                        | lmstudio-community-gemma-4-26b-a4b-it-qat-mlx-4bit | reported | 65.4 tok/s     | n=1
  m4-pro-48gb                        | mlx-community-qwen3-5-4b-mlx-8bit      | reported | 47.8 tok/s     | n=1
  m4-pro-48gb                        | unsloth-qwen3-6-35b-a3b-gguf           | reported | 47.1 tok/s     | n=1
  m4-pro-48gb                        | mlx-community-qwen3-5-9b-mlx-8bit      | reported | 27.4 tok/s     | n=1
  m4-pro-48gb                        | unsloth-qwen3-6-27b-gguf               | reported | 12 tok/s       | n=1

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