// RUN โ€” FITS

Can the NVIDIA GeForce RTX 3080 Ti run SPIDER2?

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

SPIDER2 on the NVIDIA GeForce RTX 3080 Ti

QuantFits?Max contextSpeedBeyond context
FP16 yes 146k 812822.3-968706 t/s ~35650.1-42487.1 (slow)
Q8 yes 146k 1625644.6-1937412.1 t/s ~71300.2-84974.2 (slow)
Q4 yes 146k 2902936.8-3459664.5 t/s ~127321.8-151739.7 (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 NVIDIA GeForce RTX 3080 Ti run SPIDER2?โ–ถ

Yes. At Q4 it generates ~2902936.8-3459664.5 tok/s and fits up to 146k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~127321.8-151739.7 tok/s.

How much context fits?โ–ถ

At Q4, up to about 146k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.