🔌
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 CARP?
Yes — here's how fast and up to what context, with real measured numbers.
CARP on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1715k | 155-184.7 t/s | ~5.7-6.8 (slow) |
| Q8 | yes | 1725k | 309.9-369.4 t/s | ~11.4-13.5 (slow) |
| Q4 | yes | 1729k | 553.5-659.6 t/s | ~20.3-24.2 (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 CARP?▶
Yes. At Q4 it generates ~553.5-659.6 tok/s and fits up to 1729k context; beyond that it spills to SSD swap and slows to ~20.3-24.2 tok/s.
How much context fits?▶
At Q4, up to about 1729k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.