// RUN โ€” FITS

Can the Jetson Thor T2000 run TurboVLA?

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

TurboVLA on the Jetson Thor T2000

QuantFits?Max contextSpeedBeyond context
FP16 yes 191k 250-298 t/s ~18.3-21.8 (slow)
Q8 yes 194k 500.1-596 t/s ~36.5-43.5 (slow)
Q4 yes 195k 892.9-1064.2 t/s ~65.2-77.7 (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 TurboVLA?โ–ถ

Yes. At Q4 it generates ~892.9-1064.2 tok/s and fits up to 195k context; beyond that it spills to SSD swap and slows to ~65.2-77.7 tok/s.

How much context fits?โ–ถ

At Q4, up to about 195k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.