🔌
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 RMBG v2.0?
Yes — here's how fast and up to what context, with real measured numbers.
RMBG v2.0 on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1728k | 450.9-537.4 t/s | ~16.5-19.7 (slow) |
| Q8 | yes | 1732k | 901.8-1074.7 t/s | ~33-39.4 (slow) |
| Q4 | yes | 1733k | 1610.3-1919.1 t/s | ~59-70.3 (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 RMBG v2.0?▶
Yes. At Q4 it generates ~1610.3-1919.1 tok/s and fits up to 1733k context; beyond that it spills to SSD swap and slows to ~59-70.3 tok/s.
How much context fits?▶
At Q4, up to about 1733k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.