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=ryzen-ai-max-395-64gb.
Full machine surface: /llms.txt · REST API: api.bestmodel.run

  rig                               | model                                  | basis    | value          | n
  ryzen-ai-max-395-64gb              | unsloth-gemma-4-e2b-it-gguf            | measured | 105.6 tok/s    | n=4
  ryzen-ai-max-395-64gb              | poolside-laguna-xs-2-1-gguf            | reported | 92.4 tok/s     | n=2
  ryzen-ai-max-395-64gb              | unsloth-qwen3-30b-a3b-gguf             | reported | 89 tok/s       | n=2
  ryzen-ai-max-395-64gb              | unsloth-qwen3-coder-30b-a3b-instruct-gguf | measured | 88.9 tok/s     | n=3
  ryzen-ai-max-395-64gb              | openai-gpt-oss-20b                     | reported | 75.2 tok/s     | n=2
  ryzen-ai-max-395-64gb              | internscience-agents-a1-q4-k-m-gguf    | measured | 74.6 tok/s     | n=6
  ryzen-ai-max-395-64gb              | samuelcardillo-qwopus-moe-35b-a3b-gguf | measured | 74.6 tok/s     | n=4
  ryzen-ai-max-395-64gb              | unsloth-phi-4-mini-reasoning-gguf      | reported | 73.9 tok/s     | n=2
  ryzen-ai-max-395-64gb              | unsloth-glm-4-7-flash-gguf             | measured | 65.7 tok/s     | n=4
  ryzen-ai-max-395-64gb              | qwen-qwen3-6-35b-a3b                   | measured | 62.4 tok/s     | n=11
  ryzen-ai-max-395-64gb              | ornith-ai-ornith-1-0-35b               | measured | 60.6 tok/s     | n=3
  ryzen-ai-max-395-64gb              | ornith-ai-ornith-1-0-35b-gguf          | measured | 60.6 tok/s     | n=6
  ryzen-ai-max-395-64gb              | unsloth-qwen3-6-35b-a3b-gguf           | reported | 60.5 tok/s     | n=1
  ryzen-ai-max-395-64gb              | unsloth-qwen3-6-35b-a3b-mtp-gguf       | measured | 60.5 tok/s     | n=3
  ryzen-ai-max-395-64gb              | unsloth-qwen-agentworld-35b-a3b-gguf   | measured | 60.5 tok/s     | n=6
  ryzen-ai-max-395-64gb              | unsloth-gemma-4-26b-a4b-it-gguf        | reported | 49.2 tok/s     | n=2
  ryzen-ai-max-395-64gb              | poolside-laguna-s-2-1                  | measured | 43.7 tok/s     | n=8
  ryzen-ai-max-395-64gb              | unsloth-qwen3-8b-gguf                  | reported | 40.9 tok/s     | n=2
  ryzen-ai-max-395-64gb              | unsloth-qwen3-5-9b                     | reported | 38.1 tok/s     | n=2
  ryzen-ai-max-395-64gb              | ornith-ai-ornith-1-0-9b-gguf           | reported | 35.5 tok/s     | n=2
  ryzen-ai-max-395-64gb              | poolside-laguna-s-2-1                  | measured | 32.8 tok/s     | n=7
  ryzen-ai-max-395-64gb              | unsloth-laguna-s-2-1-gguf              | reported | 32.6 tok/s     | n=1
  ryzen-ai-max-395-64gb              | qwen-qwen3-coder-30b-a3b-instruct      | measured | 31.9 tok/s     | n=8
  ryzen-ai-max-395-64gb              | stepfun-ai-step-3-7-flash              | measured | 30.8 tok/s     | n=5
  ryzen-ai-max-395-64gb              | stepfun-ai-step-3-7-flash-gguf         | measured | 30.8 tok/s     | n=5
  ryzen-ai-max-395-64gb              | zai-org-glm-4-5-air                    | reported | 29.6 tok/s     | n=1
  ryzen-ai-max-395-64gb              | unsloth-qwen3-6-27b-mtp-gguf           | measured | 29.2 tok/s     | n=3
  ryzen-ai-max-395-64gb              | google-gemma-3-12b-it                  | reported | 24.6 tok/s     | n=2
  ryzen-ai-max-395-64gb              | unsloth-qwen3-8-27b-gguf               | measured | 24.1 tok/s     | n=8
  ryzen-ai-max-395-64gb              | bartowski-deepseek-r1-distill-qwen-14b-gguf | reported | 24 tok/s       | n=2
  ryzen-ai-max-395-64gb              | qwen-qwen3-8-27b                       | measured | 18.5 tok/s     | n=6
  ryzen-ai-max-395-64gb              | qwen-qwen3-8-27b                       | measured | 17.3 tok/s     | n=5
  ryzen-ai-max-395-64gb              | meta-models-muse-glimmer-30b-gguf      | measured | 12.9 tok/s     | n=6
  ryzen-ai-max-395-64gb              | meta-models-muse-glimmer-30b           | reported | 12.9 tok/s     | n=2
  ryzen-ai-max-395-64gb              | qwen-qwen3-6-27b                       | measured | 12.4 tok/s     | n=6
  ryzen-ai-max-395-64gb              | qwen-qwen3-6-27b                       | reported | 7.7 tok/s      | n=1

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