// RUN โ€” FITS

Can the AMD Instinct MI300X 192GB run ResNet-RS?

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

ResNet-RS on the AMD Instinct MI300X 192GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 2667k 8695.3-10627.6 t/s ~65.6-80.2 (slow)
Q8 yes 2669k 17390.6-21255.2 t/s ~131.3-160.4 (slow)
Q4 yes 2671k 31054.7-37955.7 t/s ~234.4-286.5 (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 AMD Instinct MI300X 192GB run ResNet-RS?โ–ถ

Yes. At Q4 it generates ~31054.7-37955.7 tok/s and fits up to 2671k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~234.4-286.5 tok/s.

How much context fits?โ–ถ

At Q4, up to about 2671k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.