Models /Granite Speech 3.3 8b /Can I run it?

Compatibility check

Can I run Granite Speech 3.3 8b on Apple M3 Ultra (80-core GPU)?

Yes — it runs.best quant:

Decode speed

110 tok/sest

Usable context

262K

Memory

512 GB

Fits in memory110 tok/sest · 262K ctx

Fits in memory94 tok/sest · 262K ctx

Fits in memory63 tok/sest · 262K ctx

What is quantisation? →

Granite Speech 3.3 8b

8.6B · full specs & benchmarks →

Apple M3 Ultra (80-core GPU)

512 GB · everything it runs →