// RUN โ€” FITS

Can the Jetson Thor T2000 run T5Gemma (Gemma 9B-9B)?

Yes โ€” here's how fast and up to what context, with real measured numbers.

T5Gemma (Gemma 9B-9B) on the Jetson Thor T2000

QuantFits?Max contextSpeedBeyond context
FP16 no โ€” โ€” โ€”
Q8 no โ€” โ€” โ€”
Q4 yes 14k 10.7-12.7 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 T2000 run T5Gemma (Gemma 9B-9B)?โ–ถ

Yes. At Q4 it generates ~10.7-12.7 tok/s and fits up to 14k context; beyond that it spills to SSD swap and slows to ~0.8-0.9 tok/s.

How much context fits?โ–ถ

At Q4, up to about 14k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.