🔌
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 OLMo-1B?
Yes — here's how fast and up to what context, with real measured numbers.
OLMo-1B on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1704k | 99.6-118.8 t/s | ~3.7-4.4 (slow) |
| Q8 | yes | 1720k | 199.3-237.5 t/s | ~7.3-8.7 (slow) |
| Q4 | yes | 1726k | 355.9-424.1 t/s | ~13-15.5 (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 OLMo-1B?▶
Yes. At Q4 it generates ~355.9-424.1 tok/s and fits up to 1726k context; beyond that it spills to SSD swap and slows to ~13-15.5 tok/s.
How much context fits?▶
At Q4, up to about 1726k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.