community pool / model answer

Qwen3Guard-Gen-4B

chat · Qwen/Qwen3Guard-Gen-4B

66.8 tok/s · model median · 2 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RX 9060 XT 16GB8-bit66.8 tok/sreported2,2645.5 GB8,192llama.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.