Upgrade to latest CUDA on Jetson without upgrading JetPack

Starting from CUDA 11.8, Jetson users on JetPack 5.0 and above will have the ability to upgrade to the latest and greatest CUDA versions without the need to update the JetPack version or Jetson Linux BSP.

To know more on how to upgrade CUDA on your Jetson device today, read our blog Simplifying CUDA Upgrades for NVIDIA Jetson Users | NVIDIA Technical Blog

  • On the CUDA 11.8 Downloads page, download the CUDA installer for “aarch64-Jetson” and follow the installation instructions to upgrade your Jetson device to CUDA 11.8.
  • For more details on CUDA upgradable package on Jetson, refer to the CUDA for Tegra App Note.
  • For information on all the new features that CUDA 11.8 brings in, refer to the release notes.
  • Do register for our deep-dive webinar here where the CUDA and Jetson teams will walk you through details on this new feature and get an opportunity to ask questions live! “

Now with cuda 12.2. Driver is too old…

import torch
Traceback (most recent call last):
File “”, line 1, in
AttributeError: module ‘torch.cuda’ has no attribute ‘is_avaible’. Did you mean: ‘is_available’?
/home/jetson/miniconda3/envs/prueba/lib/python3.11/site-packages/torch/cuda/init.py:138: UserWarning: CUDA initialization: The NVIDIA driver on your system is too old (found version 11040). Please update your GPU driver by downloading and installing a new version from the URL: Official Drivers | NVIDIA Alternatively, go to: https://pytorch.org to install a PyTorch version that has been compiled with your version of the CUDA driver. (Triggered internally at /home/jetson/Projects/pytorch/c10/cuda/CUDAFunctions.cpp:108.)
return torch._C._cuda_getDeviceCount() > 0