// RUN — FITS

Can the Apple M3 Max 36GB run AWD-LSTM + MoS + Partial Shuffled?

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

AWD-LSTM + MoS + Partial Shuffled on the Apple M3 Max 36GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 470k 2914.3-3514.3 t/s ~97.1-117.1 (slow)
Q8 yes 471k 5828.6-7028.6 t/s ~194.3-234.3 (slow)
Q4 yes 471k 10408.2-12551 t/s ~346.9-418.4 (slow)

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

→ All models on the Apple M3 Max 36GB → AWD-LSTM + MoS + Partial Shuffled specs → Other hardware

FAQ

Can the Apple M3 Max 36GB run AWD-LSTM + MoS + Partial Shuffled?▶

Yes. At Q4 it generates ~10408.2-12551 tok/s and fits up to 471k context; beyond that it spills to SSD swap and slows to ~346.9-418.4 tok/s.

How much context fits?▶

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