Compatibility check
Can I run Llama Guard 4 12B on Apple M4 Pro (20-core GPU)?
Yes — it runs.best quant:
Decode speed
26 tok/sest
Usable context
262K
Memory
64 GB
Fits in memory26 tok/sest · 262K ctx
Fits in memory22 tok/sest · 262K ctx
Fits in memory15 tok/sest · 262K ctx