// RUN โ€” FITS

Can the NVIDIA GeForce RTX 4070 Super run SRU++ Large?

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

SRU++ Large on the NVIDIA GeForce RTX 4070 Super

QuantFits?Max contextSpeedBeyond context
FP16 yes 138k 786.2-936.9 t/s ~62.4-74.4 (slow)
Q8 yes 142k 1572.3-1873.8 t/s ~124.8-148.7 (slow)
Q4 yes 144k 2807.7-3346.2 t/s ~222.8-265.6 (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 GeForce RTX 4070 Super run SRU++ Large?โ–ถ

Yes. At Q4 it generates ~2807.7-3346.2 tok/s and fits up to 144k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~222.8-265.6 tok/s.

How much context fits?โ–ถ

At Q4, up to about 144k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.