🔌
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 GemNet-T (OC20)?
Yes — here's how fast and up to what context, with real measured numbers.
GemNet-T (OC20) on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1735k | 52444.7-62502.6 t/s | ~1921.1-2289.5 (slow) |
| Q8 | yes | 1735k | 104889.5-125005.3 t/s | ~3842.1-4578.9 (slow) |
| Q4 | yes | 1735k | 187302.6-223223.7 t/s | ~6860.9-8176.7 (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 GemNet-T (OC20)?▶
Yes. At Q4 it generates ~187302.6-223223.7 tok/s and fits up to 1735k context; beyond that it spills to SSD swap and slows to ~6860.9-8176.7 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.