// RUN โ€” FITS

Can the Intel Arc Pro B70 run GPT-1?

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

GPT-1 on the Intel Arc Pro B70

QuantFits?Max contextSpeedBeyond context
FP16 yes 423k 1507-1870.8 t/s ~99.1-123.1 (slow)
Q8 yes 425k 3014-3741.5 t/s ~198.3-246.2 (slow)
Q4 yes 425k 5382.2-6681.3 t/s ~354.1-439.6 (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 B70 run GPT-1?โ–ถ

Yes. At Q4 it generates ~5382.2-6681.3 tok/s and fits up to 425k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~354.1-439.6 tok/s.

How much context fits?โ–ถ

At Q4, up to about 425k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.