🔌
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 MatterGen?
Yes — here's how fast and up to what context, with real measured numbers.
MatterGen on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1733k | 2129.2-2537.5 t/s | ~78-92.9 (slow) |
| Q8 | yes | 1734k | 4258.3-5075 t/s | ~156-185.9 (slow) |
| Q4 | yes | 1735k | 7604.2-9062.5 t/s | ~278.5-332 (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 MatterGen?▶
Yes. At Q4 it generates ~7604.2-9062.5 tok/s and fits up to 1735k context; beyond that it spills to SSD swap and slows to ~278.5-332 tok/s.
How much context fits?▶
At Q4, up to about 1735k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.