// RUN โ€” FITS

Can the Intel Arc Pro B50 run EXAONE Deep 7.8B?

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

EXAONE Deep 7.8B on the Intel Arc Pro B50

QuantFits?Max contextSpeedBeyond context
FP16 no โ€” โ€” โ€”
Q8 yes 41k 16.7-20.7 t/s ~3-3.7 (slow)
Q4 yes 68k 29.7-36.9 t/s ~5.3-6.6 (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 EXAONE Deep 7.8B?โ–ถ

Yes. At Q4 it generates ~29.7-36.9 tok/s and fits up to 68k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~5.3-6.6 tok/s.

How much context fits?โ–ถ

At Q4, up to about 68k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.