// RUN โ€” FITS

Can the Instinct MI430X run SANA 1.6B?

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

SANA 1.6B on the Instinct MI430X

QuantFits?Max contextSpeedBeyond context
FP16 yes 5991k 4453.6-5443.3 t/s ~7.6-9.3 (slow)
Q8 yes 6016k 8907.2-10886.5 t/s ~15.3-18.7 (slow)
Q4 yes 6027k 15905.6-19440.2 t/s ~27.3-33.4 (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 MI430X run SANA 1.6B?โ–ถ

Yes. At Q4 it generates ~15905.6-19440.2 tok/s and fits up to 6027k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~27.3-33.4 tok/s.

How much context fits?โ–ถ

At Q4, up to about 6027k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.