// RUN โ€” FITS

Can the Intel Arc Pro B50 run mBART-50?

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

mBART-50 on the Intel Arc Pro B50

QuantFits?Max contextSpeedBeyond context
FP16 yes 183k 106.5-132.2 t/s ~19-23.6 (slow)
Q8 yes 192k 213-264.4 t/s ~38-47.2 (slow)
Q4 yes 197k 380.3-472.1 t/s ~67.9-84.3 (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 Pro B50 run mBART-50?โ–ถ

Yes. At Q4 it generates ~380.3-472.1 tok/s and fits up to 197k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~67.9-84.3 tok/s.

How much context fits?โ–ถ

At Q4, up to about 197k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.