// RUN โ€” FITS

Can the NVIDIA GeForce RTX 3090 run ESM2-35M?

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

ESM2-35M on the NVIDIA GeForce RTX 3090

QuantFits?Max contextSpeedBeyond context
FP16 yes 313k 9761.1-11633.1 t/s ~417.1-497.1 (slow)
Q8 yes 313k 19522.3-23266.3 t/s ~834.3-994.3 (slow)
Q4 yes 314k 34861.2-41546.9 t/s ~1489.8-1775.5 (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 NVIDIA GeForce RTX 3090 run ESM2-35M?โ–ถ

Yes. At Q4 it generates ~34861.2-41546.9 tok/s and fits up to 314k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1489.8-1775.5 tok/s.

How much context fits?โ–ถ

At Q4, up to about 314k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.