Qwen3.6 35B (3B active)

User reports Qwen3.6 35B-A3B at 90 tok/s on an M4 Pro Mac mini with 48GB unified memory. Setup is oMLX 0.7.0 with a custom oQ4 quantised model using MTP headers. User notes performance had recently fallen as low as 10-11 tok/s before this improvement.