// RUN โ€” FITS

Can the Apple M2 Ultra 192GB run Mixtral 8x7B?

Yes โ€” here's how fast and up to what context, with real measured numbers.

Mixtral 8x7B on the Apple M2 Ultra 192GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 238k 5.8-7 t/s ~0.1-0.1 (slow)
Q8 yes 380k 11.6-14 t/s ~0.1-0.2 (slow)
Q4 yes 443k 20.8-25.1 t/s ~0.3-0.3 (slow)

Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.

FAQ

Can the Apple M2 Ultra 192GB run Mixtral 8x7B?โ–ถ

Yes. At Q4 it generates ~20.8-25.1 tok/s and fits up to 443k context; beyond that it spills to SSD swap and slows to ~0.3-0.3 tok/s.

How much context fits?โ–ถ

At Q4, up to about 443k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.