🔌
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 Minitron 8B?
Yes — here's how fast and up to what context, with real measured numbers.
Minitron 8B on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 741k | 12-14.3 t/s | ~0.4-0.5 (slow) |
| Q8 | yes | 804k | 24-28.6 t/s | ~0.9-1 (slow) |
| Q4 | yes | 832k | 42.9-51.1 t/s | ~1.6-1.9 (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 Minitron 8B?▶
Yes. At Q4 it generates ~42.9-51.1 tok/s and fits up to 832k context; beyond that it spills to SSD swap and slows to ~1.6-1.9 tok/s.
How much context fits?▶
At Q4, up to about 832k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.