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 DMPFold?
Yes — here's how fast and up to what context, with real measured numbers.
DMPFold on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2672k | 768421.1-915789.5 t/s | ~3842.1-4578.9 (slow) |
| Q8 | yes | 2672k | 1536842.1-1831578.9 t/s | ~7684.2-9157.9 (slow) |
| Q4 | yes | 2672k | 2744360.9-3270676.7 t/s | ~13721.8-16353.4 (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 DMPFold?▶
Yes. At Q4 it generates ~2744360.9-3270676.7 tok/s and fits up to 2672k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~13721.8-16353.4 tok/s.
How much context fits?▶
At Q4, up to about 2672k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.