🔌
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 AWD-LSTM-DOC (fin) (23M)?
Yes — here's how fast and up to what context, with real measured numbers.
AWD-LSTM-DOC (fin) (23M) on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1734k | 4332.4-5163.3 t/s | ~158.7-189.1 (slow) |
| Q8 | yes | 1735k | 8664.8-10326.5 t/s | ~317.4-378.3 (slow) |
| Q4 | yes | 1735k | 15472.8-18440.2 t/s | ~566.8-675.5 (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 AWD-LSTM-DOC (fin) (23M)?▶
Yes. At Q4 it generates ~15472.8-18440.2 tok/s and fits up to 1735k context; beyond that it spills to SSD swap and slows to ~566.8-675.5 tok/s.
How much context fits?▶
At Q4, up to about 1735k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.