community pool / model answer

Qwen3.6-35B-A3B-Quark-W8A8-INT8

chat · nameistoken/Qwen3.6-35B-A3B-Quark-W8A8-INT8

93.6 tok/s · model median · 3 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Arc Pro B70 32GB ×28-bit85.9 tok/sreported273- GB32,768vllm
Arc Pro B70 32GB ×48-bit96.5 tok/sreported132- GB32,768vllm

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.