🔌
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 LDM-1.45B?
Yes — here's how fast and up to what context, with real measured numbers.
LDM-1.45B on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1912k | 1208.3-1440 t/s | ~10.1-12 (slow) |
| Q8 | yes | 1934k | 2416.6-2880 t/s | ~20.1-24 (slow) |
| Q4 | yes | 1944k | 4315.3-5142.9 t/s | ~36-42.9 (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 LDM-1.45B?▶
Yes. At Q4 it generates ~4315.3-5142.9 tok/s and fits up to 1944k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~36-42.9 tok/s.
How much context fits?▶
At Q4, up to about 1944k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.