Compatibility check
Can I run Rocinante 12B on Apple M3 Ultra (80-core GPU)?
Yes — it runs.best quant:
Decode speed
78 tok/sest
Usable context
66K
Memory
512 GB
Fits in memory78 tok/sest · 66K ctx
Fits in memory66 tok/sest · 66K ctx
Fits in memory44 tok/sest · 66K ctx
