🔌
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 LUKE?
Yes — here's how fast and up to what context, with real measured numbers.
LUKE on the Jetson Thor T3000
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 402k | 206.3-245.9 t/s | ~7.6-9 (slow) |
| Q8 | yes | 409k | 412.6-491.7 t/s | ~15.1-18 (slow) |
| Q4 | yes | 412k | 736.8-878.1 t/s | ~27-32.2 (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 T3000 run LUKE?▶
Yes. At Q4 it generates ~736.8-878.1 tok/s and fits up to 412k context; beyond that it spills to SSD swap and slows to ~27-32.2 tok/s.
How much context fits?▶
At Q4, up to about 412k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.