this is my device info
(base) user@user:~/pi0/openpi$ sudo jetson_release
Software part of jetson-stats 4.3.2 - (c) 2024, Raffaello Bonghi
Model: NVIDIA Jetson AGX Thor Developer Kit - Jetpack 7.0 [L4T 38.2.2]
NV Power Mode[0]: MAXN
Serial Number: [XXX Show with: jetson_release -s XXX]
Hardware:
P-Number: p3834-0008
Module: Not available
Platform:
Distribution: Ubuntu 24.04 Noble Numbat
Release: 6.8.12-rt-tegra
jtop:
Version: 4.3.2
Service: Active
Libraries:
CUDA: 13.0.48
cuDNN: 9.12.0.46
TensorRT: 10.13.3.9
VPI: 4.0.0~er5
Vulkan: 1.4.304
OpenCV: 4.8.0 - with CUDA: NO
the pi0 model and pi05 model depends on cuda12.x, which is different from thr cuda13.0 on AGX Thor device
kayccc
March 4, 2026, 5:21am
4
Moving to Jetson Thor forum.
Hi,
Please stay on JetPack 7.0/7.1 with CUDA 13.0.
We have a tutorial with pi below for Thor:
Thanks.
thanks, I have complete the pi0.5 model inference through pytorch on Jetson Thor device without docker. I try to use JAX to run this model,which I have complete on Jetson (AGX/NX) Orin device. How can I do thin on Jetson Thor device. So the problem is “How to build jax-0.5.3 on jetson Thor device“
Hi,
Let’s dicusss the JAX issue on the below topic instead:
I try to build jax0.5.3 on jetson thor device, but the code did not recognize the sm_110
Thanks.
system
Closed
March 23, 2026, 7:47am
9
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.