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 SRU++ Large?
Yes — here's how fast and up to what context, with real measured numbers.
SRU++ Large on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1949k | 7487.2-8923.1 t/s | ~62.4-74.4 (slow) |
| Q8 | yes | 1953k | 14974.4-17846.2 t/s | ~124.8-148.7 (slow) |
| Q4 | yes | 1954k | 26739.9-31868.1 t/s | ~222.8-265.6 (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 H200 run SRU++ Large?▶
Yes. At Q4 it generates ~26739.9-31868.1 tok/s and fits up to 1954k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~222.8-265.6 tok/s.
How much context fits?▶
At Q4, up to about 1954k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.