No GPU support in PyTorch for JetPack 6.2 (L4T R36.4) – missing .whl file

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

Hi NVIDIA team,

I’m using JetPack 6.2 (L4T R36.4.3) on a Jetson Orin Nano and trying to install PyTorch with GPU support. I noticed in other forum that there’s no official .whl available yet for JetPack 6.2 on the NVIDIA PyTorch repository.

I followed a community suggestion and ran the following command:

pip3 install torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 --index-url https://pypi.jetson-ai-lab.dev/jp6/cu126

However, this resulted in a connection error and ultimately failed with:

WARNING: Retrying (Retry(total=4, ...): Failed to establish a new connection: [Errno 113] No route to host ... ERROR: Could not find a version that satisfies the requirement torch==2.8.0 (from versions: none) ERROR: No matching distribution found for torch==2.8.0

This suggests the index is either not public or not functional.

❗ The issue

JetPack 6.2 ships with CUDA 12.6 and cuDNN 8.9.5, but there is no official PyTorch GPU .whl for this version. This makes it hard to use JetPack 6.2 for real-time AI.

❓ My questions

  1. Will NVIDIA release official PyTorch .whl files with GPU support for JetPack 6.2 (L4T R36.4)?
  2. Is there an estimated timeline?
  3. Are there recommended alternatives or Docker images for this version?

📋 System Information

  • Device: Jetson Orin Nano
  • JetPack: 6.2
  • L4T: R36.4.3
  • CUDA: 12.6
  • cuDNN: 8.9.5
  • Python: 3.10
  • Ubuntu: 20.04

Thank you very much for your help.

3

Hi,

The URL has been changed.

Please try the package in the link below:

Thanks.

4

Thanks to the instructions shared, I was able to get PyTorch running with GPU support on my Jetson Orin Nano using JetPack 6.2 + CUDA 12.6.

Here’s what worked for me:

pip uninstall torch torchvision torchaudio -y pip install --upgrade pip setuptools wheel  pip install torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 \ --index-url https://pypi.jetson-ai-lab.io/jp6/cu126

Then I verified it with:

python -c "import torch; print(torch.__version__, torch.version.cuda, torch.cuda.is_available(), torch.cuda.get_device_name(0))"

Output:

2.8.0 12.6 True NVIDIA Orin
5

@AastaLLL the url https://pypi.jetson-ai-lab.io/ is currently down. Could you provide a new one? Thanks.

6

Hi,

Please try it again. It should work now.
We will check with our internal team to improve the stability of the server.

Thanks, and sorry for the inconvenience.

1 Like
7

Hi @AastaLLL, the url seems to be down again. the .dev mirror is also down. Is there any other way to get the wheel file? I need torch and torchvision for the same purposes as the original poster. Thank you!

8

Hi,

Please check the .io URL again.
It was unstable due to the server upgrade, but it should be much better now.

Thanks, and sorry for the inconvenience.

9

Dear Nvidia. Is there some page somewhere when the current commands to install important packages like PyTorch are located or is the standard procedure to just post a new topic each time you want to install it to get the current commands?

1 Like
10

Hi,

You can find the important update for the package in the repo below:

These packages are compiled with the source code listed in the above repo.
Thanks.

11

I tried to install this version of torch: [torch-2.8.0-cp312-cp312-linux_aarch64.whl] on my Jetson Orin Nano, jetpack 6.2, and CUDA 12.9 and I got this error message. I want the GPU capabilities not just CPU:

pip install torch-2.8.0-cp312-cp312-linux_aarch64.whl
ERROR: torch-2.8.0-cp312-cp312-linux_aarch64.whl is not a supported wheel on this platform.

12

Hi,

It looks like you are trying to install a Python3.12 package.
Have you maually upgrade it to the 3.12?

Do you upgrade CUDA 12.9 manually?
If you are using the default CUDA 12.6, you can find the packages for python 3.10 below:

Thanks.

13

Hi!
I am currently using python 3.10, but I am trying to also use pyrealsense2 2.56.5, should I downgrade my pyrealsense2? No I did not upgrade CUDA 12.9 manually, if i was to try and do it manually what are the correct steps to ensure that it is not the problem?"

14

Hi @AastaLLL

Looks like the link https://pypi.jetson-ai-lab.io/jp6/cu126 still seems to be down,
need to install torch, torchvision, torchaudio
is there any alternative to the above link?

1 Like
15

Hi,

Sorry that due to some system issues, the server is not available now.

Our internal team is actively working on the server issue.
Will keep you updated on the latest status.

Thanks.

16

Hi, all

The server is up.
Please try it again. Sorry for the inconvenience.

Thanks.

17

The URL is down again. Can you check please?

18

From my location (US) the links is (has been) working as of 8/8/25:

jetson-ai-lab
jetson-ai-lab814×415 17.9 KB
19

Hi,

Please check if you are using the new .io URL.

Thanks.