// RUN โ€” FITS

Can the NVIDIA GeForce RTX 3070 Ti run SGPT BE 5.8B?

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

SGPT BE 5.8B on the NVIDIA GeForce RTX 3070 Ti

QuantFits?Max contextSpeedBeyond context
FP16 no โ€” โ€” โ€”
Q8 yes 457 76.5-91.2 t/s ~5-6 (slow)
Q4 yes 20k 136.7-162.9 t/s ~9-10.7 (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 3070 Ti run SGPT BE 5.8B?โ–ถ

Yes. At Q4 it generates ~136.7-162.9 tok/s and fits up to 20k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~9-10.7 tok/s.

How much context fits?โ–ถ

At Q4, up to about 20k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.