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-3080-12gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  rtx-3080-12gb                      | liquidai-lfm2-5-8b-a1b-gguf            | reported | 328.4 tok/s    | n=1
  rtx-3080-12gb                      | liquidai-lfm2-5-8b-a1b-gguf            | reported | 277.8 tok/s    | n=1
  rtx-3080-12gb                      | deepseek-ai-deepseek-r1-distill-qwen-7b | reported | 144 tok/s      | n=1
  rtx-3080-12gb                      | qwen-qwen2-5-coder-7b-instruct         | measured | 140.8 tok/s    | n=3
  rtx-3080-12gb                      | meta-llama-llama-3-1-8b-instruct       | reported | 134 tok/s      | n=2
  rtx-3080-12gb                      | google-gemma-4-e4b-it                  | measured | 130 tok/s      | n=3
  rtx-3080-12gb                      | nvidia-nvidia-nemotron-nano-9b-v2      | reported | 98 tok/s       | n=1
  rtx-3080-12gb                      | ornith-ai-ornith-1-5-9b-gguf           | reported | 92 tok/s       | n=1
  rtx-3080-12gb                      | deepseek-ai-deepseek-coder-v2-lite-instruct | reported | 87 tok/s       | n=2
  rtx-3080-12gb                      | ornith-ai-ornith-1-5-9b-gguf           | reported | 82.8 tok/s     | n=1
  rtx-3080-12gb                      | microsoft-phi-4                        | reported | 77 tok/s       | n=1
  rtx-3080-12gb                      | google-gemma-3n-e4b-it                 | reported | 75 tok/s       | n=2
  rtx-3080-12gb                      | qwen-qwen3-14b                         | reported | 75 tok/s       | n=1
  rtx-3080-12gb                      | prism-ml-bonsai-27b-gguf               | reported | 72.9 tok/s     | n=1
  rtx-3080-12gb                      | qwen-qwen2-5-coder-14b-instruct        | measured | 72 tok/s       | n=5
  rtx-3080-12gb                      | google-gemma-4-12b-it-qat-q4-0-gguf    | reported | 71.2 tok/s     | n=2
  rtx-3080-12gb                      | ornith-ai-ornith-1-5-35b-a3b-gguf      | reported | 60.8 tok/s     | n=2
  rtx-3080-12gb                      | orcarouter-qwen3-8-27b-uncensored-gguf | reported | 60.7 tok/s     | n=2
  rtx-3080-12gb                      | agentica-org-deepcoder-14b-preview     | reported | 59 tok/s       | n=1
  rtx-3080-12gb                      | ggml-org-gpt-oss-20b-gguf              | reported | 57.1 tok/s     | n=1
  rtx-3080-12gb                      | google-gemma-3-12b-it                  | measured | 51.5 tok/s     | n=4
  rtx-3080-12gb                      | unsloth-qwen3-8-27b-gguf               | measured | 44.4 tok/s     | n=3
  rtx-3080-12gb                      | qwen-qwen2-5-14b-instruct              | measured | 44 tok/s       | n=4
  rtx-3080-12gb                      | qwen-qwen3-6-35b-a3b                   | measured | 35.4 tok/s     | n=5
  rtx-3080-12gb                      | qwen-qwen3-30b-a3b                     | reported | 31 tok/s       | n=1
  rtx-3080-12gb                      | openai-gpt-oss-20b                     | reported | 21 tok/s       | n=1
  rtx-3080-12gb                      | unsloth-muse-glimmer-30b-gguf          | reported | 13.8 tok/s     | n=1
  rtx-3080-12gb                      | meta-models-muse-glimmer-30b-gguf      | reported | 6 tok/s        | n=1

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