Compatibility check
Can I run Mistral Small 3 on Apple M5 Max (32-core GPU)?
Yes — it runs.best quant:
Decode speed
23 tok/sest
Usable context
33K
Memory
64 GB
Fits in memory23 tok/sest · 33K ctx
Fits in memory19 tok/sest · 33K ctx
Fits in memory13 tok/sest · 33K ctx