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-5070-12gb. Full machine surface: /llms.txt · REST API: api.bestmodel.run rig | model | basis | value | n rtx-5070-12gb | jetbrains-mellum2-12b-a2-5b-thinking-gguf-q4-k-m | reported | 265.5 tok/s | n=1 rtx-5070-12gb | jetbrains-mellum2-12b-a2-5b-instruct-gguf-q4-k-m | reported | 261.4 tok/s | n=1 rtx-5070-12gb | apodex-apodex-1-1-mini | reported | 182.3 tok/s | n=2 rtx-5070-12gb | nvidia-nvidia-nemotron-3-nano-4b-gguf | measured | 182.2 tok/s | n=3 rtx-5070-12gb | google-gemma-4-e4b-it-qat-q4-0-gguf | reported | 152.7 tok/s | n=1 rtx-5070-12gb | openbmb-minicpm5-2b | reported | 118.2 tok/s | n=1 rtx-5070-12gb | ornith-ai-ornith-1-5-9b | reported | 102 tok/s | n=1 rtx-5070-12gb | ggml-org-gemma-4-12b-it-gguf | reported | 72.1 tok/s | n=1 rtx-5070-12gb | turboderp-qwen3-8-27b-exl3 | measured | 71.9 tok/s | n=4 rtx-5070-12gb | mia-ailab-gemmable-4-12b-mtp-gguf | reported | 71.2 tok/s | n=1 rtx-5070-12gb | kai-os-grug-12b-gguf | reported | 69.2 tok/s | n=1 rtx-5070-12gb | unsloth-qwen3-5-4b-gguf | reported | 68 tok/s | n=1 rtx-5070-12gb | unsloth-qwen3-5-9b-gguf | reported | 67.3 tok/s | n=1 rtx-5070-12gb | unsloth-qwen3-14b-gguf | reported | 66.6 tok/s | n=1 rtx-5070-12gb | unsloth-qwen3-6-35b-a3b-mtp-gguf | reported | 64.5 tok/s | n=1 rtx-5070-12gb | unsloth-gemma-4-12b-it-gguf | reported | 62.8 tok/s | n=1 rtx-5070-12gb | unsloth-gemma-4-12b-it-gguf | reported | 51.3 tok/s | n=1 rtx-5070-12gb | unsloth-qwen3-8-27b-gguf | measured | 46.3 tok/s | n=5 rtx-5070-12gb | atomicchat-qwen3-8-27b-gguf | reported | 46 tok/s | n=2 Capture or correct a number: /console (signed, Ed25519).