// RUN โ€” FITS

Can the NVIDIA DGX Spark run Stable Diffusion 3 Medium?

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

Stable Diffusion 3 Medium on the NVIDIA DGX Spark

QuantFits?Max contextSpeedBeyond context
FP16 yes 1659k 39.9-47.5 t/s ~1.5-1.7 (slow)
Q8 yes 1697k 79.7-95 t/s ~2.9-3.5 (slow)
Q4 yes 1714k 142.4-169.7 t/s ~5.2-6.2 (slow)

Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.

FAQ

Can the NVIDIA DGX Spark run Stable Diffusion 3 Medium?โ–ถ

Yes. At Q4 it generates ~142.4-169.7 tok/s and fits up to 1714k context; beyond that it spills to SSD swap and slows to ~5.2-6.2 tok/s.

How much context fits?โ–ถ

At Q4, up to about 1714k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.