Compatibility check
Can I run Mixtral 8x22B Instruct on Apple M2 Max (38-core GPU)?
Partially — with CPU offload.best quant:
88.7 GB of weights, plus 4.3 GB for the software that runs it and the smallest conversation it can hold, comes to 93 GB against the 72 GB this 96 GB device leaves free.
Decode speed
—
Usable context
—
Memory
96 GB
Spills to system RAM
Too largeest
Too large
Plan B — rent it
Mixtral 8x22B Instruct is hosted by 2 providers from $6.00 per 1M output tokens (Mistral AI).Compare providers →
