New: connect Claude & other AIs to GenAIList over MCP โ research the catalog and contribute to the shared knowledge base. Learn how โ
Can the NVIDIA H200 run ESM1-85M?
Yes โ here's how fast and up to what context, with real measured numbers.
ESM1-85M on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1954k | 20587.5-24535.8 t/s | ~171.6-204.5 (slow) |
| Q8 | yes | 1955k | 41175.1-49071.7 t/s | ~343.1-408.9 (slow) |
| Q4 | yes | 1956k | 73526.9-87628 t/s | ~612.7-730.2 (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 H200 run ESM1-85M?โถ
Yes. At Q4 it generates ~73526.9-87628 tok/s and fits up to 1956k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~612.7-730.2 tok/s.
How much context fits?โถ
At Q4, up to about 1956k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.