Models /MOSS Transcribe Preview 2B /Can I run it?

Compatibility check

Can I run MOSS Transcribe Preview 2B on Apple M4 Max (40-core GPU)?

Yes — it runs.best quant:

Decode speed

263 tok/sest

Usable context

33K

Memory

128 GB

Fits in memory263 tok/sest · 33K ctx

Fits in memory224 tok/sest · 33K ctx

Fits in memory150 tok/sest · 33K ctx

What is quantisation? →

MOSS Transcribe Preview 2B

2.4B · full specs & benchmarks →

Apple M4 Max (40-core GPU)

128 GB · everything it runs →