🔌
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 OPT-6.7B?
Yes — here's how fast and up to what context, with real measured numbers.
OPT-6.7B on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 876k | 261.5-311.6 t/s | ~2.2-2.6 (slow) |
| Q8 | yes | 927k | 523-623.3 t/s | ~4.4-5.2 (slow) |
| Q4 | yes | 950k | 933.9-1113 t/s | ~7.8-9.3 (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 H200 run OPT-6.7B?▶
Yes. At Q4 it generates ~933.9-1113 tok/s and fits up to 950k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~7.8-9.3 tok/s.
How much context fits?▶
At Q4, up to about 950k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.