community pool / model answer

Qwen3-8B-GGUF

chat · Qwen/Qwen3-8B-GGUF

56.6 tok/s · model median · 7 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
M3 Ultra 512GB4-bit88.3 tok/sreported5,802- GB8,192llama.cpp
RTX 3060 12GB ×24-bit53.8 tok/smeasured1,8947.1 GB32,768llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.