🔌
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 SRU++ Large?
Yes — here's how fast and up to what context, with real measured numbers.
SRU++ Large on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1728k | 425.8-507.5 t/s | ~15.6-18.6 (slow) |
| Q8 | yes | 1731k | 851.7-1015 t/s | ~31.2-37.2 (slow) |
| Q4 | yes | 1733k | 1520.8-1812.5 t/s | ~55.7-66.4 (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 SRU++ Large?▶
Yes. At Q4 it generates ~1520.8-1812.5 tok/s and fits up to 1733k context; beyond that it spills to SSD swap and slows to ~55.7-66.4 tok/s.
How much context fits?▶
At Q4, up to about 1733k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.