Compatibility Between JetPack 5.1.3, CUDA 12.6, and PyTorch: Finding the Right Version

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
1

I would like to know for the Jetpack version 5.1.3 and the version of CUDA 12.6 which version of Pytorch fits so that it works correctly between them and where can I install it?

3

Hi,

The latest CUDA for JetPack 5 environment is 12.2.
If you want 12.6, please upgrade to JetPack 6.1 and you can find the compatible PyTorch package below:

http://jetson.webredirect.org/jp6/cu126

Thanks.

1 Like
4

Thank you!
How can I upgrade the jetpack version 5.1.3 to jetpack 6.1.0? For jetpack 6.1.0 and CUDA 12.6 the right version of PyTorch is torch-2.5.0?

5

Hi,

Please use SDKmanager to set up the environment.
And our prebuilt PyTorch for the JetPack 6.1 users is v2.5.0.

Thanks.

1 Like
6

Hi, I want to download and install Libtorch for Pytorch version 2.4, Python version 3.10 and CUDA version 12.2, but I don’t know how can I do?

7

Hi,

For CUDA 12.2, please setup with JetPack 6.0 GA.
And then you can find the compatible PyTorch/TorchVision in the link below:

(The latest version is 2.3.0)

Thanks.

9

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.