// RUN โ€” FITS

Can the NVIDIA H200 run Granite 3.0 2B?

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

Granite 3.0 2B on the NVIDIA H200

QuantFits?Max contextSpeedBeyond context
FP16 yes 1880k 700.8-835.2 t/s ~5.8-7 (slow)
Q8 yes 1918k 1401.6-1670.4 t/s ~11.7-13.9 (slow)
Q4 yes 1935k 2502.9-2982.9 t/s ~20.9-24.9 (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 Granite 3.0 2B?โ–ถ

Yes. At Q4 it generates ~2502.9-2982.9 tok/s and fits up to 1935k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~20.9-24.9 tok/s.

How much context fits?โ–ถ

At Q4, up to about 1935k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.