// RUN โ€” FITS

Can the NVIDIA L40S 48GB run Distilled Grandmaster?

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

Distilled Grandmaster on the NVIDIA L40S 48GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 643k 1168-1392 t/s ~54.1-64.4 (slow)
Q8 yes 647k 2336-2784 t/s ~108.1-128.9 (slow)
Q4 yes 649k 4171.4-4971.4 t/s ~193.1-230.2 (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 NVIDIA L40S 48GB run Distilled Grandmaster?โ–ถ

Yes. At Q4 it generates ~4171.4-4971.4 tok/s and fits up to 649k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~193.1-230.2 tok/s.

How much context fits?โ–ถ

At Q4, up to about 649k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.