// RUN โ€” FITS

Can the NVIDIA H100 PCIe 80GB run Minerva 1B?

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

Minerva 1B on the NVIDIA H100 PCIe 80GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 1070k 730-870 t/s ~14.6-17.4 (slow)
Q8 yes 1085k 1460-1740 t/s ~29.2-34.8 (slow)
Q4 yes 1092k 2607.1-3107.1 t/s ~52.1-62.1 (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 H100 PCIe 80GB run Minerva 1B?โ–ถ

Yes. At Q4 it generates ~2607.1-3107.1 tok/s and fits up to 1092k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~52.1-62.1 tok/s.

How much context fits?โ–ถ

At Q4, up to about 1092k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.