// RUN โ€” FITS

Can the Intel Arc A770 16GB run AWD-LSTM + MoS + Partial Shuffled?

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

AWD-LSTM + MoS + Partial Shuffled on the Intel Arc A770 16GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 201k 4640-5760 t/s ~331.4-411.4 (slow)
Q8 yes 201k 9280-11520 t/s ~662.9-822.9 (slow)
Q4 yes 201k 16571.4-20571.4 t/s ~1183.7-1469.4 (slow)

Estimate (memory-bound). Beyond "max context" the KV cache spills to system RAM offload (depends on your system RAM) and speed drops to the "beyond" figure.

FAQ

Can the Intel Arc A770 16GB run AWD-LSTM + MoS + Partial Shuffled?โ–ถ

Yes. At Q4 it generates ~16571.4-20571.4 tok/s and fits up to 201k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1183.7-1469.4 tok/s.

How much context fits?โ–ถ

At Q4, up to about 201k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.