🔌
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 BLOOM-560M?
Yes — here's how fast and up to what context, with real measured numbers.
BLOOM-560M on the Jetson Thor T2000
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 180k | 89.3-106.4 t/s | ~6.5-7.8 (slow) |
| Q8 | yes | 188k | 178.6-212.8 t/s | ~13-15.5 (slow) |
| Q4 | yes | 192k | 318.9-380.1 t/s | ~23.3-27.7 (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 BLOOM-560M?▶
Yes. At Q4 it generates ~318.9-380.1 tok/s and fits up to 192k context; beyond that it spills to SSD swap and slows to ~23.3-27.7 tok/s.
How much context fits?▶
At Q4, up to about 192k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.