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=gtx-1080-ti-11gb. Full machine surface: /llms.txt · REST API: api.bestmodel.run rig | model | basis | value | n gtx-1080-ti-11gb | mistralai-ministral-3-3b-instruct-2512 | reported | 94.9 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-5-9b | reported | 64.9 tok/s | n=2 gtx-1080-ti-11gb | qwen-qwen2-5-7b-instruct | reported | 57.6 tok/s | n=1 gtx-1080-ti-11gb | nousresearch-hermes-3-llama-3-1-8b | reported | 53.1 tok/s | n=1 gtx-1080-ti-11gb | meta-llama-llama-3-1-8b-instruct | reported | 52.8 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-8b | reported | 50.7 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen2-5-7b-instruct | reported | 41.1 tok/s | n=1 gtx-1080-ti-11gb | unsloth-qwen3-6-35b-a3b-mtp-gguf | reported | 39.4 tok/s | n=1 gtx-1080-ti-11gb | nousresearch-hermes-3-llama-3-1-8b | reported | 38.2 tok/s | n=1 gtx-1080-ti-11gb | meta-llama-llama-3-1-8b-instruct | reported | 38.1 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-8b | reported | 36.8 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-5-9b | reported | 34.3 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-6-35b-a3b | measured | 21.6 tok/s | n=5 gtx-1080-ti-11gb | qwen-qwen3-6-35b-a3b | reported | 14.6 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-6-35b-a3b | reported | 10.3 tok/s | n=2 gtx-1080-ti-11gb | qwen-qwen3-5-35b-a3b | reported | 9.3 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-6-35b-a3b | reported | 6.4 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-5-122b-a10b | reported | 3.2 tok/s | n=1 gtx-1080-ti-11gb | qwen-qwen3-5-27b | reported | 2.5 tok/s | n=2 Capture or correct a number: /console (signed, Ed25519).