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 ProtGPT2?
Yes — here's how fast and up to what context, with real measured numbers.
ProtGPT2 on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2650k | 3956.6-4715.4 t/s | ~19.8-23.6 (slow) |
| Q8 | yes | 2661k | 7913.3-9430.9 t/s | ~39.6-47.2 (slow) |
| Q4 | yes | 2666k | 14130.9-16840.9 t/s | ~70.7-84.2 (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 ProtGPT2?▶
Yes. At Q4 it generates ~14130.9-16840.9 tok/s and fits up to 2666k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~70.7-84.2 tok/s.
How much context fits?▶
At Q4, up to about 2666k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.