Compatibility check
Can I run Granite Speech 3.3 2b on Apple M1 (8-core GPU)?
Yes — it runs.best quant:
Decode speed
26 tok/sest
Usable context
66K
Memory
16 GB
Fits in memory26 tok/sest · 66K ctx
Fits in memory22 tok/sest · 66K ctx
Fits in memory15 tok/sest · 66K ctx