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=rtx3060-rtx2080-pooled-20gb-20gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  rtx3060-rtx2080-pooled-20gb-20gb   | deepseek-ai-deepseek-coder-v2-lite-instruct | reported | 118.5 tok/s    | n=2
  rtx3060-rtx2080-pooled-20gb-20gb   | deepseek-ai-deepseek-r1-distill-qwen-7b | reported | 106.5 tok/s    | n=2
  rtx3060-rtx2080-pooled-20gb-20gb   | qwen-qwen2-5-coder-7b-instruct         | reported | 106 tok/s      | n=1
  rtx3060-rtx2080-pooled-20gb-20gb   | meta-llama-llama-3-1-8b-instruct       | reported | 105 tok/s      | n=2
  rtx3060-rtx2080-pooled-20gb-20gb   | qwen-qwen3-30b-a3b                     | reported | 91 tok/s       | n=1
  rtx3060-rtx2080-pooled-20gb-20gb   | openai-gpt-oss-20b                     | reported | 77 tok/s       | n=1
  rtx3060-rtx2080-pooled-20gb-20gb   | google-gemma-4-e4b-it                  | measured | 76 tok/s       | n=3
  rtx3060-rtx2080-pooled-20gb-20gb   | google-gemma-3n-e4b-it                 | reported | 74 tok/s       | n=1
  rtx3060-rtx2080-pooled-20gb-20gb   | google-gemma-3-12b-it                  | measured | 44 tok/s       | n=3
  rtx3060-rtx2080-pooled-20gb-20gb   | microsoft-phi-4                        | reported | 36 tok/s       | n=1
  rtx3060-rtx2080-pooled-20gb-20gb   | qwen-qwen2-5-14b-instruct              | measured | 36 tok/s       | n=3
  rtx3060-rtx2080-pooled-20gb-20gb   | agentica-org-deepcoder-14b-preview     | reported | 35 tok/s       | n=1
  rtx3060-rtx2080-pooled-20gb-20gb   | qwen-qwen2-5-coder-14b-instruct        | measured | 35 tok/s       | n=3
  rtx3060-rtx2080-pooled-20gb-20gb   | qwen-qwen3-14b                         | reported | 35 tok/s       | n=1

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