🔌
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 CogVLM-17B?
Yes — here's how fast and up to what context, with real measured numbers.
CogVLM-17B on the Jetson Thor T3000
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | no | — | — | — |
| Q8 | yes | 39k | 11.7-14 t/s | ~0.4-0.5 (slow) |
| Q4 | yes | 68k | 20.9-24.9 t/s | ~0.8-0.9 (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 CogVLM-17B?▶
Yes. At Q4 it generates ~20.9-24.9 tok/s and fits up to 68k context; beyond that it spills to SSD swap and slows to ~0.8-0.9 tok/s.
How much context fits?▶
At Q4, up to about 68k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.