🔌
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 Intern-S2-Preview?
Yes — here's how fast and up to what context, with real measured numbers.
Intern-S2-Preview on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 401k | 83.4-99.4 t/s | ~0.4-0.5 (slow) |
| Q8 | yes | 535k | 166.9-198.9 t/s | ~0.8-1 (slow) |
| Q4 | yes | 593k | 298-355.1 t/s | ~1.5-1.8 (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 Intern-S2-Preview?▶
Yes. At Q4 it generates ~298-355.1 tok/s and fits up to 593k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1.5-1.8 tok/s.
How much context fits?▶
At Q4, up to about 593k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.