// RUN โ€” FITS

Can the Apple M4 Max 36GB run AlphaLink?

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

AlphaLink on the Apple M4 Max 36GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 469k 1996.1-2407.1 t/s ~36.6-44.1 (slow)
Q8 yes 470k 3992.3-4814.2 t/s ~73.1-88.2 (slow)
Q4 yes 471k 7129-8596.8 t/s ~130.6-157.5 (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 M4 Max 36GB run AlphaLink?โ–ถ

Yes. At Q4 it generates ~7129-8596.8 tok/s and fits up to 471k context; beyond that it spills to SSD swap and slows to ~130.6-157.5 tok/s.

How much context fits?โ–ถ

At Q4, up to about 471k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.