community pool / model answer

Qwen3.8-27B-INT8-W8A16-MTP

chat · lued/Qwen3.8-27B-INT8-W8A16-MTP

161.1 tok/s · model median · 6 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
CMP 170HX 64GB8-bit161.1 tok/smeasured5,31954.6 GB65,536vllm

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.