Compatibility check
Can I run Qwen3 VL 8B Instruct on Apple M4 Pro (20-core GPU)?
Yes — it runs.best quant:
Decode speed
36 tok/sest
Usable context
262K
Memory
64 GB
Fits in memory36 tok/sest · 262K ctx
Fits in memory31 tok/sest · 262K ctx
Fits in memory20 tok/sest · 131K ctx