Compatibility check
Can I run Hunyuan A13B Instruct on Apple M5 Max (40-core GPU)?
Yes — it runs.best quant:
Decode speed
46 tok/sest
Usable context
131K
Memory
128 GB
Fits in memory46 tok/sest · 131K ctx
Fits in memory40 tok/sest · 131K ctx
Fits in memoryest26 tok/sest · 16K ctx