New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA B200 run FoundationStereo?
Yes — here's how fast and up to what context, with real measured numbers.
FoundationStereo on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2662k | 8708.6-10378.8 t/s | ~43.5-51.9 (slow) |
| Q8 | yes | 2667k | 17417.2-20757.5 t/s | ~87.1-103.8 (slow) |
| Q4 | yes | 2670k | 31102.2-37067 t/s | ~155.5-185.3 (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 FoundationStereo?▶
Yes. At Q4 it generates ~31102.2-37067 tok/s and fits up to 2670k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~155.5-185.3 tok/s.
How much context fits?▶
At Q4, up to about 2670k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.