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-3060-12gb-x2. Full machine surface: /llms.txt · REST API: api.bestmodel.run rig | model | basis | value | n rtx-3060-12gb-x2 | liquidai-lfm2-5-1-2b-thinking-gguf | reported | 251.2 tok/s | n=1 rtx-3060-12gb-x2 | unsloth-lfm2-8b-a1b-gguf | measured | 231.4 tok/s | n=8 rtx-3060-12gb-x2 | liquidai-lfm2-5-8b-a1b-gguf | measured | 218.2 tok/s | n=6 rtx-3060-12gb-x2 | liquidai-lfm2-8b-a1b-gguf | reported | 193.5 tok/s | n=1 rtx-3060-12gb-x2 | liquidai-lfm2-5-8b-a1b-gguf | reported | 175.6 tok/s | n=1 rtx-3060-12gb-x2 | qwen-qwen3-0-6b-gguf | measured | 144.4 tok/s | n=8 rtx-3060-12gb-x2 | openbmb-minicpm5-1b-gguf | reported | 125.9 tok/s | n=1 rtx-3060-12gb-x2 | qwen-qwen3-1-7b-gguf | measured | 112.6 tok/s | n=8 rtx-3060-12gb-x2 | arcee-ai-trinity-mini | measured | 99.3 tok/s | n=16 rtx-3060-12gb-x2 | unsloth-qwen3-6-35b-a3b-mtp-gguf | reported | 98.9 tok/s | n=1 rtx-3060-12gb-x2 | unsloth-qwen3-30b-a3b-thinking-2507-gguf | reported | 94.3 tok/s | n=1 rtx-3060-12gb-x2 | unsloth-gpt-oss-20b-gguf | reported | 93.4 tok/s | n=1 rtx-3060-12gb-x2 | internscience-agents-a1 | measured | 91.7 tok/s | n=7 rtx-3060-12gb-x2 | internscience-agents-a1-q4-k-m-gguf | measured | 91.7 tok/s | n=7 rtx-3060-12gb-x2 | jackrong-qwopus3-6-35b-a3b-v1-mtp-gguf | measured | 91.6 tok/s | n=6 rtx-3060-12gb-x2 | coherelabs-north-mini-code-1-0 | measured | 89.5 tok/s | n=6 rtx-3060-12gb-x2 | unsloth-gemma-4-26b-a4b-it-qat-gguf | measured | 87.8 tok/s | n=21 rtx-3060-12gb-x2 | unsloth-nemotron-3-nano-30b-a3b-gguf | reported | 86.2 tok/s | n=1 rtx-3060-12gb-x2 | ornith-ai-ornith-1-0-35b-gguf | measured | 85.9 tok/s | n=7 rtx-3060-12gb-x2 | cyankiwi-qwen3-5-9b-awq-4bit | measured | 79.1 tok/s | n=9 rtx-3060-12gb-x2 | unsloth-qwen-agentworld-35b-a3b-gguf | measured | 78.3 tok/s | n=11 rtx-3060-12gb-x2 | unsloth-qwen3-6-35b-a3b-gguf | measured | 78.2 tok/s | n=9 rtx-3060-12gb-x2 | mradermacher-gemma-4-19b-a4b-it-reap-gguf | measured | 77.5 tok/s | n=6 rtx-3060-12gb-x2 | unsloth-phi-4-mini-reasoning-gguf | measured | 73.3 tok/s | n=8 rtx-3060-12gb-x2 | lmstudio-community-gemma-4-26b-a4b-it-gguf | reported | 69.5 tok/s | n=1 rtx-3060-12gb-x2 | unsloth-gemma-4-e4b-it-gguf | measured | 69.3 tok/s | n=17 rtx-3060-12gb-x2 | unsloth-gemma-4-26b-a4b-it-gguf | measured | 68.2 tok/s | n=14 rtx-3060-12gb-x2 | qwen-qwen3-6-35b-a3b | reported | 67.6 tok/s | n=1 rtx-3060-12gb-x2 | lmstudio-community-gemma-4-e4b-it-gguf | reported | 62.1 tok/s | n=1 rtx-3060-12gb-x2 | lmstudio-community-openreasoning-nemotron-7b-gguf | measured | 59.8 tok/s | n=8 rtx-3060-12gb-x2 | bartowski-ui-tars-7b-dpo-gguf | measured | 59.7 tok/s | n=8 rtx-3060-12gb-x2 | lmstudio-community-acereason-nemotron-1-1-7b-gguf | measured | 59.7 tok/s | n=8 rtx-3060-12gb-x2 | bartowski-deepseek-r1-distill-qwen-7b-gguf | measured | 59.7 tok/s | n=8 rtx-3060-12gb-x2 | dhruvallabs-webworld-8b-gguf | measured | 54 tok/s | n=6 rtx-3060-12gb-x2 | qwen-qwen3-8b-gguf | measured | 53.8 tok/s | n=6 rtx-3060-12gb-x2 | empero-ai-qwythos-9b-v2-gguf | measured | 53.2 tok/s | n=17 rtx-3060-12gb-x2 | adriabama06-qwen3-5-4b-q8-0-am17an-mtp | reported | 52.4 tok/s | n=1 rtx-3060-12gb-x2 | kai-os-carnice-9b-gguf | measured | 51.6 tok/s | n=9 rtx-3060-12gb-x2 | unsloth-qwen3-5-9b-gguf | measured | 51 tok/s | n=9 rtx-3060-12gb-x2 | bartowski-internlm3-8b-instruct-gguf | measured | 50.9 tok/s | n=8 rtx-3060-12gb-x2 | nvidia-nemotron-orchestrator-8b | measured | 47.5 tok/s | n=8 rtx-3060-12gb-x2 | bartowski-nvidia-nemotron-cascade-8b-gguf | measured | 47.5 tok/s | n=8 rtx-3060-12gb-x2 | unsloth-qwen3-5-9b-mtp-gguf | reported | 47.3 tok/s | n=1 rtx-3060-12gb-x2 | weiboai-vibethinker-3b | measured | 45.7 tok/s | n=9 rtx-3060-12gb-x2 | nvidia-nvidia-nemotron-nano-9b-v2 | measured | 45.2 tok/s | n=8 rtx-3060-12gb-x2 | nvidia-nvidia-nemotron-nano-12b-v2 | measured | 39 tok/s | n=8 rtx-3060-12gb-x2 | unsloth-muse-glimmer-30b-gguf | reported | 37.8 tok/s | n=2 rtx-3060-12gb-x2 | jgebbeken-gemma-4-coder-gguf | measured | 37.1 tok/s | n=6 rtx-3060-12gb-x2 | google-gemma-3-12b-it | measured | 34.3 tok/s | n=9 rtx-3060-12gb-x2 | unsloth-phi-4-gguf | reported | 34.2 tok/s | n=1 rtx-3060-12gb-x2 | dhruvallabs-webworld-14b-gguf | measured | 32.3 tok/s | n=6 rtx-3060-12gb-x2 | unsloth-qwen3-14b-gguf | measured | 29.7 tok/s | n=8 rtx-3060-12gb-x2 | nvidia-nemotron-cascade-14b-thinking | measured | 29.7 tok/s | n=8 rtx-3060-12gb-x2 | prism-ml-bonsai-27b-gguf | measured | 29.3 tok/s | n=18 rtx-3060-12gb-x2 | bartowski-deepseek-r1-distill-qwen-14b-gguf | measured | 28.3 tok/s | n=8 rtx-3060-12gb-x2 | microsoft-phi-4-reasoning | measured | 28.1 tok/s | n=16 rtx-3060-12gb-x2 | rekaai-reka-flash-3 | measured | 22.3 tok/s | n=14 rtx-3060-12gb-x2 | bartowski-mistralai-mistral-small-3-2-24b-instruct-2506-gguf | reported | 21.8 tok/s | n=1 rtx-3060-12gb-x2 | nousresearch-deephermes-3-mistral-24b-preview | measured | 21.5 tok/s | n=8 rtx-3060-12gb-x2 | unsloth-qwen3-8-27b-gguf | measured | 18.7 tok/s | n=5 Capture or correct a number: /console (signed, Ed25519).