🔌
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 GPT-2 (774M)?
Yes — here's how fast and up to what context, with real measured numbers.
GPT-2 (774M) on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1933k | 2263.6-2697.7 t/s | ~18.9-22.5 (slow) |
| Q8 | yes | 1945k | 4527.1-5395.3 t/s | ~37.7-45 (slow) |
| Q4 | yes | 1950k | 8084.2-9634.6 t/s | ~67.4-80.3 (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 GPT-2 (774M)?▶
Yes. At Q4 it generates ~8084.2-9634.6 tok/s and fits up to 1950k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~67.4-80.3 tok/s.
How much context fits?▶
At Q4, up to about 1950k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.