Compatibility check
Can I run Hunyuan A13B Instruct on Apple M2 Ultra (76-core GPU)?
Yes — it runs.best quant:
Decode speed
60 tok/sest
Usable context
131K
Memory
192 GB
Fits in memory60 tok/sest · 131K ctx
Fits in memory52 tok/sest · 131K ctx
Fits in memory35 tok/sest · 131K ctx