Compatibility check
Can I run Llama 3 8B Lunaris on Apple M4 Pro (20-core GPU)?
Yes — it runs.best quant:
Decode speed
39 tok/sest
Usable context
8K
Memory
64 GB
Fits in memory39 tok/sest · 8K ctx
Fits in memory34 tok/sest · 8K ctx
Fits in memory23 tok/sest · 8K ctx