🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the NVIDIA H200 run Unified-IO (XL)?
Yes — here's how fast and up to what context, with real measured numbers.
Unified-IO (XL) on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1867k | 599-713.8 t/s | ~5-5.9 (slow) |
| Q8 | yes | 1912k | 1197.9-1427.7 t/s | ~10-11.9 (slow) |
| Q4 | yes | 1931k | 2139.2-2549.5 t/s | ~17.8-21.2 (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 Unified-IO (XL)?▶
Yes. At Q4 it generates ~2139.2-2549.5 tok/s and fits up to 1931k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~17.8-21.2 tok/s.
How much context fits?▶
At Q4, up to about 1931k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.