🔌
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 Zagreus 0.4B?
Yes — here's how fast and up to what context, with real measured numbers.
Zagreus 0.4B on the Jetson Thor T3000
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 404k | 249.1-296.9 t/s | ~9.1-10.9 (slow) |
| Q8 | yes | 410k | 498.2-593.8 t/s | ~18.3-21.8 (slow) |
| Q4 | yes | 413k | 889.7-1060.3 t/s | ~32.6-38.8 (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 Zagreus 0.4B?▶
Yes. At Q4 it generates ~889.7-1060.3 tok/s and fits up to 413k context; beyond that it spills to SSD swap and slows to ~32.6-38.8 tok/s.
How much context fits?▶
At Q4, up to about 413k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.