🔌
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 MoEFold2D?
Yes — here's how fast and up to what context, with real measured numbers.
MoEFold2D on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1735k | 103796.9-123703.1 t/s | ~3802.1-4531.3 (slow) |
| Q8 | yes | 1735k | 207593.8-247406.3 t/s | ~7604.2-9062.5 (slow) |
| Q4 | yes | 1735k | 370703.1-441796.9 t/s | ~13578.9-16183 (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 MoEFold2D?▶
Yes. At Q4 it generates ~370703.1-441796.9 tok/s and fits up to 1735k context; beyond that it spills to SSD swap and slows to ~13578.9-16183 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.