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=m5-pro-64gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  m5-pro-64gb                        | mlx-community-qwen3-6-35b-a3b-mxfp4    | reported | 106.1 tok/s    | n=1
  m5-pro-64gb                        | mlx-community-qwen3-5-35b-a3b-4bit     | reported | 104.7 tok/s    | n=1
  m5-pro-64gb                        | mlx-community-qwen3-6-35b-a3b-4bit     | reported | 103.9 tok/s    | n=1
  m5-pro-64gb                        | mlx-community-qwen3-coder-30b-a3b-instruct-4bit | reported | 101.1 tok/s    | n=1
  m5-pro-64gb                        | mlx-community-qwen3-6-35b-a3b-4bit-dwq | reported | 78.9 tok/s     | n=1
  m5-pro-64gb                        | mlx-community-glm-4-7-flash-4bit       | reported | 77.5 tok/s     | n=1
  m5-pro-64gb                        | mlx-community-qwen3-coder-next-4bit    | reported | 76.5 tok/s     | n=1
  m5-pro-64gb                        | mlx-community-qwen3-6-35b-a3b-8bit     | reported | 67.3 tok/s     | n=1
  m5-pro-64gb                        | mlx-community-qwen3-6-27b-mxfp4        | reported | 19.1 tok/s     | n=1
  m5-pro-64gb                        | mlx-community-qwen3-6-27b-4bit         | reported | 18.1 tok/s     | n=1
  m5-pro-64gb                        | unsloth-qwen3-6-27b-gguf               | reported | 11.4 tok/s     | n=1
  m5-pro-64gb                        | unsloth-qwen3-6-27b-mlx-8bit           | reported | 8.4 tok/s      | n=1
  m5-pro-64gb                        | mlx-community-qwen3-6-27b-bf16         | reported | 5.4 tok/s      | n=1

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