// RUN โ€” FITS

Can the Apple M2 Max 64GB run DistilProtBert?

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

DistilProtBert on the Apple M2 Max 64GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 849k 591.3-713 t/s ~14.8-17.8 (slow)
Q8 yes 853k 1182.6-1426.1 t/s ~29.6-35.7 (slow)
Q4 yes 854k 2111.8-2546.6 t/s ~52.8-63.7 (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 Max 64GB run DistilProtBert?โ–ถ

Yes. At Q4 it generates ~2111.8-2546.6 tok/s and fits up to 854k context; beyond that it spills to SSD swap and slows to ~52.8-63.7 tok/s.

How much context fits?โ–ถ

At Q4, up to about 854k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.