🔌
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 B200 run Unified-IO (XL)?
Yes — here's how fast and up to what context, with real measured numbers.
Unified-IO (XL) on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2583k | 998.3-1189.7 t/s | ~5-5.9 (slow) |
| Q8 | yes | 2628k | 1996.6-2379.5 t/s | ~10-11.9 (slow) |
| Q4 | yes | 2647k | 3565.3-4249.1 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 B200 run Unified-IO (XL)?▶
Yes. At Q4 it generates ~3565.3-4249.1 tok/s and fits up to 2647k 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 2647k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.