New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA H200 run OpenELM-450M?
Yes — here's how fast and up to what context, with real measured numbers.
OpenELM-450M on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1943k | 3893.3-4640 t/s | ~32.4-38.7 (slow) |
| Q8 | yes | 1950k | 7786.7-9280 t/s | ~64.9-77.3 (slow) |
| Q4 | yes | 1953k | 13904.8-16571.4 t/s | ~115.9-138.1 (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 OpenELM-450M?▶
Yes. At Q4 it generates ~13904.8-16571.4 tok/s and fits up to 1953k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~115.9-138.1 tok/s.
How much context fits?▶
At Q4, up to about 1953k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.