// RUN โ€” FITS

Can the Intel Arc Pro B60 run Spec-Drafter?

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

Spec-Drafter on the Intel Arc Pro B60

QuantFits?Max contextSpeedBeyond context
FP16 yes 299k 264.5-328.3 t/s ~23.2-28.8 (slow)
Q8 yes 306k 529-656.6 t/s ~46.4-57.6 (slow)
Q4 yes 310k 944.6-1172.6 t/s ~82.9-102.9 (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 B60 run Spec-Drafter?โ–ถ

Yes. At Q4 it generates ~944.6-1172.6 tok/s and fits up to 310k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~82.9-102.9 tok/s.

How much context fits?โ–ถ

At Q4, up to about 310k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.