🔌
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 H200 run NTele-R1-32B-V1?
Yes — here's how fast and up to what context, with real measured numbers.
NTele-R1-32B-V1 on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 239k | 53.4-63.7 t/s | ~0.4-0.5 (slow) |
| Q8 | yes | 364k | 106.8-127.3 t/s | ~0.9-1.1 (slow) |
| Q4 | yes | 419k | 190.8-227.4 t/s | ~1.6-1.9 (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 H200 run NTele-R1-32B-V1?▶
Yes. At Q4 it generates ~190.8-227.4 tok/s and fits up to 419k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1.6-1.9 tok/s.
How much context fits?▶
At Q4, up to about 419k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.