Compatibility check
Can I run Qwen3.5-9B on Apple M4 Pro (20-core GPU)?
Yes — it runs.best quant:
Decode speed
33 tok/sest
Usable context
262K
Memory
64 GB
Fits in memory33 tok/sest · 262K ctx
Fits in memory28 tok/sest · 262K ctx
Fits in memory19 tok/sest · 262K ctx