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=m1-max-64gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  m1-max-64gb                        | openai-gpt-oss-20b                     | reported | 83.6 tok/s     | n=1
  m1-max-64gb                        | unsloth-gemma-4-26b-a4b-it-qat-gguf    | reported | 76.4 tok/s     | n=1
  m1-max-64gb                        | ggml-org-gpt-oss-20b-gguf              | reported | 74.6 tok/s     | n=1
  m1-max-64gb                        | qwen-qwen3-coder-30b-a3b-instruct      | reported | 72.3 tok/s     | n=1
  m1-max-64gb                        | qwen-qwen3-30b-a3b                     | reported | 62.6 tok/s     | n=1
  m1-max-64gb                        | openai-gpt-oss-20b                     | reported | 62.4 tok/s     | n=1
  m1-max-64gb                        | poolside-laguna-xs-2-1-gguf            | reported | 62.3 tok/s     | n=1
  m1-max-64gb                        | qwen-qwen3-6-35b-a3b                   | reported | 59.3 tok/s     | n=2
  m1-max-64gb                        | nvidia-nemotron-3-nano-omni-30b-a3b-reasoning-bf16 | reported | 57.9 tok/s     | n=1
  m1-max-64gb                        | bartowski-nvidia-nemotron-cascade-2-30b-a3b-gguf | reported | 57.5 tok/s     | n=1
  m1-max-64gb                        | nvidia-gemma-4-26b-a4b-nvfp4           | reported | 52.7 tok/s     | n=1
  m1-max-64gb                        | meta-llama-llama-3-2-3b-instruct       | reported | 50.9 tok/s     | n=1
  m1-max-64gb                        | google-gemma-4-26b-a4b-it              | reported | 49 tok/s       | n=2
  m1-max-64gb                        | qwen-qwen2-5-7b-instruct               | reported | 48.7 tok/s     | n=1
  m1-max-64gb                        | mistralai-mistral-nemo-instruct-2407   | reported | 41.1 tok/s     | n=1
  m1-max-64gb                        | qwen-qwen3-14b                         | reported | 24.8 tok/s     | n=1
  m1-max-64gb                        | google-gemma-4-12b-it                  | reported | 14 tok/s       | n=1
  m1-max-64gb                        | qwen-qwen3-6-27b                       | reported | 11.3 tok/s     | n=1
  m1-max-64gb                        | mistralai-mistral-small-3-2-24b-instruct-2506 | reported | 11.2 tok/s     | n=1
  m1-max-64gb                        | qwen-qwen3-32b                         | reported | 9 tok/s        | n=1
  m1-max-64gb                        | google-gemma-4-31b-it                  | reported | 8.4 tok/s      | n=1
  m1-max-64gb                        | badtheorylabs-btl-3-compact            | reported | 3.8 tok/s      | n=2

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