LLMap
Models /Phi 4 /Can I run it?
Compatibility check

Can I run Phi 4 on Apple M2 Max (38-core GPU)?

Yes — it runs.best quant:
Decode speed
31 tok/sest
Usable context
16K
Memory
96 GB
Fits in memory31 tok/sest · 16K ctx
Fits in memory27 tok/sest · 16K ctx
Fits in memory18 tok/sest · 16K ctx
Fits in memory10 tok/sest · 16K ctx
Fits in memory · 16K ctx
What is quantisation? →
Phi 4
14.7B · full specs & benchmarks →
Apple M2 Max (38-core GPU)
96 GB · everything it runs →