🔌
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 T2000 run Llama Guard?
Yes — here's how fast and up to what context, with real measured numbers.
Llama Guard on the Jetson Thor T2000
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | no | — | — | — |
| Q8 | yes | 45k | 14.3-17 t/s | ~1-1.2 (slow) |
| Q4 | yes | 69k | 25.5-30.4 t/s | ~1.9-2.2 (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 T2000 run Llama Guard?▶
Yes. At Q4 it generates ~25.5-30.4 tok/s and fits up to 69k context; beyond that it spills to SSD swap and slows to ~1.9-2.2 tok/s.
How much context fits?▶
At Q4, up to about 69k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.