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=radeon-ai-pro-r9700-32gb-x3.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-6-27b                       | reported | 213.7 tok/s    | n=1
  radeon-ai-pro-r9700-32gb-x3        | openai-gpt-oss-20b                     | measured | 152.9 tok/s    | n=5
  radeon-ai-pro-r9700-32gb-x3        | poolside-laguna-xs-2-1                 | reported | 100.3 tok/s    | n=1
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-6-35b-a3b                   | measured | 98.9 tok/s     | n=3
  radeon-ai-pro-r9700-32gb-x3        | zai-org-glm-4-7-flash                  | reported | 97.4 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | nvidia-nemotron-cascade-2-30b-a3b      | reported | 89.8 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-6-35b-a3b                   | measured | 79.2 tok/s     | n=5
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-coder-30b-a3b-instruct      | reported | 79 tok/s       | n=2
  radeon-ai-pro-r9700-32gb-x3        | hauhaucs-qwen3-6-35b-a3b-uncensored-hauhaucs-aggressive | reported | 76.3 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | kwaipilot-kat-coder-v2-5-dev           | reported | 75.8 tok/s     | n=2
  radeon-ai-pro-r9700-32gb-x3        | google-gemma-4-26b-a4b-it              | measured | 75.3 tok/s     | n=4
  radeon-ai-pro-r9700-32gb-x3        | ornith-ai-ornith-1-0-35b               | reported | 74 tok/s       | n=2
  radeon-ai-pro-r9700-32gb-x3        | openai-gpt-oss-120b                    | measured | 69.9 tok/s     | n=4
  radeon-ai-pro-r9700-32gb-x3        | agentionai-signal-3-8-27b              | measured | 69.4 tok/s     | n=3
  radeon-ai-pro-r9700-32gb-x3        | byroneverson-glm-4-9b-chat-abliterated | reported | 65.8 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-coder-next                  | measured | 64.5 tok/s     | n=4
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-vl-8b-instruct              | reported | 58.8 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-8-27b                       | measured | 52.9 tok/s     | n=17
  radeon-ai-pro-r9700-32gb-x3        | meta-models-muse-glimmer-30b           | reported | 52.8 tok/s     | n=2
  radeon-ai-pro-r9700-32gb-x3        | poolside-laguna-s-2-1                  | measured | 48.3 tok/s     | n=13
  radeon-ai-pro-r9700-32gb-x3        | inclusionai-ling-3-0-flash             | measured | 39.3 tok/s     | n=3
  radeon-ai-pro-r9700-32gb-x3        | deepseek-ai-deepseek-v4-flash-0731     | measured | 37.3 tok/s     | n=21
  radeon-ai-pro-r9700-32gb-x3        | google-gemma-4-12b-it                  | reported | 36.7 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | ornith-ai-ornith-1-5-35b-a3b           | reported | 35.4 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | xiaomimimo-mimo-v2-5                   | measured | 33.2 tok/s     | n=4
  radeon-ai-pro-r9700-32gb-x3        | jackrong-qwopus3-5-9b-coder            | reported | 31.2 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-8-flash-next                | measured | 31.1 tok/s     | n=3
  radeon-ai-pro-r9700-32gb-x3        | cerebras-minimax-m2-5-reap-139b-a10b   | measured | 30.4 tok/s     | n=5
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-8-27b                       | measured | 28.4 tok/s     | n=3
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-5-122b-a10b                 | measured | 26.5 tok/s     | n=6
  radeon-ai-pro-r9700-32gb-x3        | ornith-ai-ornith-1-0-9b                | measured | 25.7 tok/s     | n=6
  radeon-ai-pro-r9700-32gb-x3        | zai-org-glm-5-3-flash                  | measured | 25.7 tok/s     | n=5
  radeon-ai-pro-r9700-32gb-x3        | minimaxai-minimax-m2-7                 | measured | 25.2 tok/s     | n=3
  radeon-ai-pro-r9700-32gb-x3        | mistralai-devstral-small-2-24b-instruct-2512 | reported | 23.9 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-32b                         | reported | 22.8 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | meta-models-muse-glimmer-30b           | measured | 21.3 tok/s     | n=8
  radeon-ai-pro-r9700-32gb-x3        | google-gemma-4-12b-it                  | reported | 19.3 tok/s     | n=2
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-6-27b                       | measured | 17.8 tok/s     | n=19
  radeon-ai-pro-r9700-32gb-x3        | ibm-granite-granite-4-1-30b            | reported | 17.3 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | google-gemma-4-31b-it                  | measured | 14.2 tok/s     | n=4
  radeon-ai-pro-r9700-32gb-x3        | moonshotai-kimi-dev-72b                | reported | 11.4 tok/s     | n=1
  radeon-ai-pro-r9700-32gb-x3        | deepseek-ai-deepseek-v4-flash          | measured | 9.2 tok/s      | n=9
  radeon-ai-pro-r9700-32gb-x3        | qwen-qwen3-8-flash-next                | reported | 7.7 tok/s      | n=1
  radeon-ai-pro-r9700-32gb-x3        | mistralai-mistral-medium-3-5-128b      | measured | 6.4 tok/s      | n=9

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