// RUN โ€” FITS

Can the Intel Gaudi 3 run BART-IT?

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

BART-IT on the Intel Gaudi 3

QuantFits?Max contextSpeedBeyond context
FP16 yes 1770k 7664.3-9514.3 t/s ~82.9-102.9 (slow)
Q8 yes 1772k 15328.6-19028.6 t/s ~165.7-205.7 (slow)
Q4 yes 1773k 27372.4-33979.6 t/s ~295.9-367.3 (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 Gaudi 3 run BART-IT?โ–ถ

Yes. At Q4 it generates ~27372.4-33979.6 tok/s and fits up to 1773k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~295.9-367.3 tok/s.

How much context fits?โ–ถ

At Q4, up to about 1773k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.