Compatibility check
Can I run Llama 3 8B Lunaris on Apple M2 Max (38-core GPU)?
Yes — it runs.best quant:
Decode speed
58 tok/sest
Usable context
8K
Memory
96 GB
Fits in memory58 tok/sest · 8K ctx
Fits in memory49 tok/sest · 8K ctx
Fits in memory33 tok/sest · 8K ctx