🔌
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 DGX Spark run LDM-1.45B?
Yes — here's how fast and up to what context, with real measured numbers.
LDM-1.45B on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1691k | 68.7-81.9 t/s | ~2.5-3 (slow) |
| Q8 | yes | 1713k | 137.4-163.8 t/s | ~5-6 (slow) |
| Q4 | yes | 1723k | 245.4-292.5 t/s | ~9-10.7 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the NVIDIA DGX Spark run LDM-1.45B?▶
Yes. At Q4 it generates ~245.4-292.5 tok/s and fits up to 1723k context; beyond that it spills to SSD swap and slows to ~9-10.7 tok/s.
How much context fits?▶
At Q4, up to about 1723k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.