🔌
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 Jetson Thor T3000 run SANA 1.6B?
Yes — here's how fast and up to what context, with real measured numbers.
SANA 1.6B on the Jetson Thor T3000
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 366k | 60.5-72.1 t/s | ~2.2-2.6 (slow) |
| Q8 | yes | 391k | 120.9-144.1 t/s | ~4.4-5.3 (slow) |
| Q4 | yes | 402k | 215.9-257.4 t/s | ~7.9-9.4 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the Jetson Thor T3000 run SANA 1.6B?▶
Yes. At Q4 it generates ~215.9-257.4 tok/s and fits up to 402k context; beyond that it spills to SSD swap and slows to ~7.9-9.4 tok/s.
How much context fits?▶
At Q4, up to about 402k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.