// RUN โ€” FITS

Can the Instinct MI440X run CodeT5+?

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

CodeT5+ on the Instinct MI440X

QuantFits?Max contextSpeedBeyond context
FP16 yes 1388k 385.9-471.6 t/s ~0.8-1 (slow)
Q8 yes 1449k 771.8-943.3 t/s ~1.6-1.9 (slow)
Q4 yes 1476k 1378.1-1684.4 t/s ~2.8-3.4 (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 Instinct MI440X run CodeT5+?โ–ถ

Yes. At Q4 it generates ~1378.1-1684.4 tok/s and fits up to 1476k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~2.8-3.4 tok/s.

How much context fits?โ–ถ

At Q4, up to about 1476k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.