JetPack 4.4 - L4T R32.4.3 production release

We are pleased to announce the JetPack 4.4 production release supporting Jetson AGX Xavier series, Jetson Xavier NX, Jetson TX2 series, Jetson TX1, and Jetson Nano. (JetPack 4.4 replaces JetPack 4.3 as the latest production release.)

Please read an important NOTE at the end of this announcement

The JetPack 4.4 production release builds on top JetPack 4.4 Developer Preview, and includes production versions of TensorRT 7.1 and cuDNN 8.0.

Along with JetPack 4.4, we are introducing a web based Power Estimator tool to simplify creation of custom nvpmodel power profiles for Jetson.

Please refer to the JetPack Release Notes and L4T Release Notes for additional info.

JetPack 4.4 Highlights:

JetPack 4.4 components:

  • L4T R32.4.3
  • CUDA 10.2
  • cuDNN 8.0.0
  • TensorRT 7.1.3
  • VisionWorks 1.6
  • OpenCV 4.1
  • Vulkan 1.2
  • VPI 0.3 (Developer Preview)
  • Nsight Systems 2020.2
  • Nsight Graphics 2020.1
  • Nsight Compute 2019.3

Existing installations of JetPack 4.4 Developer Preview and JetPack 4.3 can be upgraded in-place to the JetPack 4.4 production release without re-flashing the device. For more information about upgrading JetPack via the Debian package management tool, please refer to the JetPack documentation here.

Download JetPack…

JetPack Release Notes…

L4T Release Notes…

Note: Jetpack/L4T upgrade using debian package management tool is now fixed and reenabled.

We had temporarily disabled the feature of upgrading JetPack or L4T using debian package management tool due to an issue we found in one of our debian packages causing the device to not boot properly. We have fixed this issue now and have reenabled this feature. We are sorry for the inconvenience caused.

1 Like

Thank you!
I have
sudo apt-get update sudo apt-get upgrade
NVIDIA Jetson Xavier
L4T 32.4.2
Ubuntu 18.04.4 LTS
Kernel Version: 4.9.140-tegra
CUDA 10.2.89
And now it’s completly crashed: impossible to re-boot after upgrading,
I’ve to re-build my Xavier from scratch JetPack 4.4.

does it have native support for booting from ssd/nvme? without sdcard at all?
does it have gen4 pci nvme option enabled in the stock kernel?
has 32.4.3 l4t ngc container been released?

Thanks for your reply.
Xavier 16Gb Module.
I’ve no way to re-boot from eMMC.
I’ve a 250Gb nvme SSD; a 32Gb SDcard, but booting is from eMMC.
I’ve to re-flash, no way.
Fortunately all my projects were on SSD disk.

May I execute from SDKManager on Host Machine for my Xavier Target only:

  • Jetson OS Image (download)
  • Flash Jetson OS
    And than can I install all SDK components, CUDA, CUDNN, TensorRT, etc… directly from my Xavier / Ubuntu using $ sudo apt-get update && sudo apt-get upgrade ?

first you would need an updated sdkmanager
second you would have to select target etc; I prefer headless method:

sudo sdkmanager --cli install --user --logintype devzone --product Jetson --version 4.4 --targetos Linux --host --target P2888-0001 --flash all --targetimagefolder /home/nvidia/Downloads/nvidia/installation_agx --downloadfolder /home/nvidia/Downloads/nvidia

the tail with folder paths can be omitted or customized, as well as the target and other values

After flashing Jetson OS only from SDKM, to bring other JePack SDK components
Please refer to below section in JetPack documentation:

you would need to do below from Jetson
sudo apt update
sudo apt install nvidia-jetpack

But if you are upgrading from JetPack 4.4 DP, you will need to upgrade L4T as you did
sudo apt update
sudo apt upgrade
and then install nvidia-jetpack again
sudo apt install nvidia-jetpack

We were able to repdroduce the issue even after following the above steps. We are looking into fixing it.
Thank you.

It’s a long Jetson-day.
I’m still working to reinstall using 4.4.
Glad to see you got my crash issue, while upgrading for others.
My best to the Xavier-Community, from Italy!

1 Like

Please, now that the apt-get install nvidia-jetpack ended
Package: nvidia-jetpack
Version: 4.4-b186
Architecture: arm64
Maintainer: NVIDIA Corporation
Installed-Size: 194

is there any further set up to make, for example I added to /home/xxx/.bashrc

export PATH=/usr/local/cuda/bin${PATH:+:${PATH}}$ 
export LD_LIBRARY_PATH=/usr/local/cuda/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

otherwise no nvcc was found.

Thanks for helping us.

Hi ToTheFuture,

What is your current status? Are you able to boot up now?

If you cannot, please dump the serial console log and share it here.

Hi @Andrey1984, yes, it is at The ML containers (l4t-ml, l4t-pytorch, l4t-tensorflow) have also been updated for R32.4.3.

Hi all,
Yes 4.4 and OS are installed and working after regular and complete boot, both headless and with HDMI display, with WiFi and Eth0 and usbC net.
I’m going to install tensorflow for 4.4 (or is it in ?)
I’ve reinstalled Bluetooth for my Apple Keyboard and Magic Mouse, Minoru usb stereo cam should work with v4l and OpenCV 4.3 should be installed for my purpose as before.
I’ve to mount my disks, as well.
All seems to work again and updated .
Thank you .
JetsonHacks should update his to get all info.
Than jetson-inference and jetson-reinforcement demos to run as working test.
Many thanks


If anyone get system stuck after running OTA update, please refer to this thread.

Please, I need to upgrade 7 Jetson Xavier, but the modules are running without internet access and I need to manual install. But I’m not finding a link to download the Jetpack 4.4 for manual install.
More specifically I need the Cudnn 8 version needed by the Tensorflow 2.2.0.

Thank you in advance.


Your issue sounds like you didn’t try sdkmanager before.

Please check the link:

Quite happy with this straight installation from SKD Manager (running on another Ubuntu 18.04)
I also change the the GUI for my Xavier board to Gnome 3 for familiarity


I can boot up the device now after step 1. However, the device stuck again after step 4.

Any clues?

Thanks in advance.

Yes, I try with sdkmanager a lot of times, but get an error:

[375:0714/] The SUID sandbox helper binary was found, but is not configured correctly. Rather than run without sandboxing I’m aborting now. You need to make sure that /opt/nvidia/sdkmanager/chrome-sandbox is owned by root and has mode 4755.

I’ll try one more time with the command line informed by Andrey1984 above.

Hi Alex,
Could you check the permission of this file with "ls -l /opt/nvidia/sdkmanager/chrome-sandbox " command? By default we don’t need manually change permission for this file. Is it possible that the /opt/nvidia/sdkmanager directory is on a file system which is NOT ext4 format?