New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA H200 run Hyena-2 153M?
Yes — here's how fast and up to what context, with real measured numbers.
Hyena-2 153M on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1952k | 11451-13647.1 t/s | ~95.4-113.7 (slow) |
| Q8 | yes | 1954k | 22902-27294.1 t/s | ~190.8-227.5 (slow) |
| Q4 | yes | 1955k | 40896.4-48739.5 t/s | ~340.8-406.2 (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 Hyena-2 153M?▶
Yes. At Q4 it generates ~40896.4-48739.5 tok/s and fits up to 1955k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~340.8-406.2 tok/s.
How much context fits?▶
At Q4, up to about 1955k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.