Compatibility check
Can I run Llama Guard 4 12B on Apple M1 Pro (16-core GPU)?
Yes — it runs.best quant:
Decode speed
19 tok/sest
Usable context
131K
Memory
32 GB
Fits in memory19 tok/sest · 131K ctx
Fits in memory16 tok/sest · 131K ctx
Fits in memory11 tok/sest · 131K ctx