// RUN โ€” FITS

Can the NVIDIA H200 run Elyza?

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

Elyza on the NVIDIA H200

QuantFits?Max contextSpeedBeyond context
FP16 yes 624k 134.8-160.6 t/s ~1.1-1.3 (slow)
Q8 yes 703k 269.5-321.2 t/s ~2.2-2.7 (slow)
Q4 yes 738k 481.3-573.6 t/s ~4-4.8 (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 H200 run Elyza?โ–ถ

Yes. At Q4 it generates ~481.3-573.6 tok/s and fits up to 738k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~4-4.8 tok/s.

How much context fits?โ–ถ

At Q4, up to about 738k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.