🔌
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 Qwen2.5-Coder (1.5B)?
Yes — here's how fast and up to what context, with real measured numbers.
Qwen2.5-Coder (1.5B) on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1688k | 64.7-77.1 t/s | ~2.4-2.8 (slow) |
| Q8 | yes | 1711k | 129.4-154.2 t/s | ~4.7-5.6 (slow) |
| Q4 | yes | 1722k | 231.1-275.4 t/s | ~8.5-10.1 (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 Qwen2.5-Coder (1.5B)?▶
Yes. At Q4 it generates ~231.1-275.4 tok/s and fits up to 1722k context; beyond that it spills to SSD swap and slows to ~8.5-10.1 tok/s.
How much context fits?▶
At Q4, up to about 1722k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.