🔌
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 ESM1-85M?
Yes — here's how fast and up to what context, with real measured numbers.
ESM1-85M on the NVIDIA DGX Spark
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1732k | 1170.9-1395.5 t/s | ~42.9-51.1 (slow) |
| Q8 | yes | 1734k | 2341.8-2791 t/s | ~85.8-102.2 (slow) |
| Q4 | yes | 1734k | 4181.8-4983.8 t/s | ~153.2-182.6 (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 ESM1-85M?▶
Yes. At Q4 it generates ~4181.8-4983.8 tok/s and fits up to 1734k context; beyond that it spills to SSD swap and slows to ~153.2-182.6 tok/s.
How much context fits?▶
At Q4, up to about 1734k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.