// RUN โ€” FITS

Can the Instinct MI440X run MiniCPM-RobotManip?

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

MiniCPM-RobotManip on the Instinct MI440X

QuantFits?Max contextSpeedBeyond context
FP16 yes 5996k 4116-5030.7 t/s ~8.4-10.3 (slow)
Q8 yes 6019k 8232-10061.3 t/s ~16.8-20.5 (slow)
Q4 yes 6029k 14700-17966.7 t/s ~30-36.7 (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 MiniCPM-RobotManip?โ–ถ

Yes. At Q4 it generates ~14700-17966.7 tok/s and fits up to 6029k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~30-36.7 tok/s.

How much context fits?โ–ถ

At Q4, up to about 6029k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.