// RUN โ€” FITS

Can the NVIDIA H100 PCIe 80GB run HyenaDNA?

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

HyenaDNA on the NVIDIA H100 PCIe 80GB

QuantFits?Max contextSpeedBeyond context
FP16 yes 1100k 110606.1-131818.2 t/s ~2212.1-2636.4 (slow)
Q8 yes 1100k 221212.1-263636.4 t/s ~4424.2-5272.7 (slow)
Q4 yes 1100k 395021.6-470779.2 t/s ~7900.4-9415.6 (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 H100 PCIe 80GB run HyenaDNA?โ–ถ

Yes. At Q4 it generates ~395021.6-470779.2 tok/s and fits up to 1100k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~7900.4-9415.6 tok/s.

How much context fits?โ–ถ

At Q4, up to about 1100k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.