NVIDIA JetPack SDK is the most comprehensive solution for building AI applications. Use the JetPack installer to flash your Jetson Developer Kit with the latest OS image, to install developer tools for both the host PC and Developer Kit, and to install the libraries and APIs, samples, and documentation needed to jumpstart your development environment.

NVIDIA JetPack is offered free of charge, but requires NVIDIA Jetson Developer Program membership. Not a member? Join the Jetson Developer Program here.


JetPack 3.2 Developer Preview includes support for the latest L4T BSP software packages for the Jetson TX2 Developer Kit. Automatically flash your Jetson Developer Kit with the latest BSPs (L4T 28.2 Developer Preview for Jetson) and install the latest software tools required to build and profile for applications for the Jetson Platform. This package contains CUDA toolkit for the host (Ubuntu) and target (Jetson TX2) platform, the latest NVIDIA Developer Tools (Tegra Graphics Debugger 2.5 and NVIDIA System Profiler 3.9), VisionWorks 1.6, cuDNN v7.0.5 RC, Multimedia API v28.2, OpenCV 3.3.1, and TensorRT 3.0 RC2.

We encourage all users to report feedback, bugs and suggestions to improve the workflow experience of future versions of JetPack. A feedback form is available here.

Please follow the available guide to Download and Install JetPack.

 Download *    Install Guide   Documentation 

Note: An Ubuntu host PC is required to run the JetPack installer. It can flash and update software on a target Jetson device, but it cannot not run directly on that device. When no target device is present, the JetPack installer can still be used to update software on the host PC.

Note: Developer Preview versions are software that are not ready for production use.

If you're looking to run JetPack in production, please download JetPack 3.1 from the links below.

Release Highlights:

  • L4T 28.2 Developer Preview
    • Kernel is pre-configured to support Docker
  • cuDNN v7.0.5 RC
    • Grouped convolution without any API changes
    • Support new CTC Loss Layer for RNNs
  • CUDA 9.0 Toolkit
    • Fine grain control for synchronization and memory sharing control with EGLImage & EGLSync
    • EGLStream Support on x86: seamless development on x86 for Jetson deployment
    • NVRTC runtime compilation library for CUDA C++
  • NVIDIA System Profiler 3.9
    • Various bug fixes and performance enhancements.
  • Tegra Graphics Debugger 2.5
    • The Batch Histogram view can be used to visualize draw call buckets, based on primitive count
    • Added theming support for the Tegra Graphics Debugger GUI
    • In the API Inspector, you can now view query object results and conditional rendering results
  • OpenCV 3.3.1
    • OpenCV version 3.3.1 replaces OpenCV4Tegra, and includes neon and multi-threading optimizations for Jetson
  • Multimedia API v28.2
    • Buffer rotation and cropping
    • Picture-in-picture support, demonstrated with new multi-camera sample
  • TensorRT 3.0 RC2
    • Support for TensorFlow models
    • Up to 15% perf/W improvement for DL applications

Learn more about JetPack 3.2

Product Features

NVIDIA JetPack SDK is the most comprehensive solution for building AI applications. Use the JetPack installer to flash your Jetson Developer Kit with the latest OS image, to install developer tools for both the host PC and Developer Kit, and to install the libraries and APIs, samples, and documentation needed to jumpstart your development environment.

OS Image

A sample file system derived from Ubuntu for Jetson Developer Kits.

Libraries

  • CUDA Toolkit for Host (Ubuntu with cross-development support)
  • CUDA Toolkit for Jetson
  • VisionWorks
  • OpenCV
  • cuDNN
  • TensorRT
  • Multimedia API

Developer Tools

  • Tegra Graphics Debugger

    A console-grade tool that allows developers to debug and profile OpenGL ES 2.0, OpenGL ES 3.0, OpenGL ES 3.1, OpenGL 4.3-4.6, enabling developers to get the most out of the Jetson platform.

  • NVIDIA System Profiler

    A multi-core CPU PC sampling profiler that provides an interactive view of captured profiling data, helping improve overall application performance.

Samples

  • Multimedia API samples
  • CUDA samples
  • VisionWorks

Documentation

  • Jetson Documentation
  • JetPack Documentation

System Requirements

Host Platform Requirements: 

  • JetPack 3.2 Developer Preview supports Linux Ubuntu 16.04 on your host PC
    • Linux Ubuntu 14.04 on your host PC will also work, except OpenCV and VisionWorks will be unavailable for installation on the host PC

Target Platform Requirements:

  • Jetson Developer Kit
  • Additional target requirements:
    • USB Micro-B cable connecting Jetson to your Linux host for flashing.
    • (Not included in the developer kit) To connect USB peripherals such as keyboard, mouse, and [optional] USB/Ethernet adapter (for network connection), a USB hub could be connected to the USB port on the Jetson system.
    • An HDMI cable connecting the HDMI port on Jetson Developer Kit, which is connected to an external HDMI display.
    • An Ethernet cable plugged into the on-board Ethernet port, which is connected to either a secondary network card on your Linux host or the same network router providing internet access for the Linux host.

Support

We encourage all users to report feedback, bugs and suggestions to improve the workflow experience of future releases of JetPack. A feedback form is available here.

JetPack 3.1 includes support for the latest software for Jetson Developer Kits. Automatically flash your Jetson Developer Kit with the latest BSP and install the latest software tools required to build and profile for applications for the Jetson Platform.

Please follow the available guide to Download and Install JetPack.

 Download *    Install Guide   Documentation 

Note: An Ubuntu host PC is required to run the JetPack installer. The JetPack installer can flash and update software on a target Jetson device, but it cannot not run directly on that device. When no target device is present, the JetPack installer can still be used to update software on the host PC.

Release Highlights:

  • New L4T Production Release 28.1
    • This 64-bit BSP (Board Support Package) has been designed to work on both Jetson TX2 and Jetson TX1
  • TensorRT 2.1
    • New Customer Layer API enables integration of novel, user-defined layers
    • Doubled Deep Learning inference performance for batch size of one
  • cuDNN v6.0
    • New fused convolution provides better performance due to faster compute in the fused kernels
    • New dilated convolution reduces the number of parameters, which results in speed up of computation for certain applications like object detection and image segmentation that require convolution followed by upscaling
  • Multimedia API v28.1
    • New functionality
      • TNRv2 (Temporal Noise Reduction)
        • High quality spatio-temporal noise reduction using GPU. Recommended for applications where low light video quality is important and GPU requirement is acceptable. Typical GPU utilization is <8.5% for 1080p30fps operation on Jetson TX1.
      • Piecewise linear WDR Support
        • ISP now supports cameras with “built-in WDR” that combine multiple exposures on-sensor and transmit the result with a piecewise linear encoding. Functionality verified using Sony IMX-185 (and reference driver is included). This feature does not include support for other WDR technologies such as DOL or spatial interleaving.
    • New samples
      • How to share CUDA buffer with v412 camera and then process color conversion (YUYV to RGB) with CUDA algorithm
      • How to render video stream (YUV) or UI (RGB) with Tegra DRM (Direct Rendering Manager), i.e. rendering support for non-X11 and lightweight display system. Tegra DRM is implemented in user-space and is compatible with standard DRM 2.0

Learn more about JetPack 3.1

Product Features

JetPack includes host (Ubuntu Desktop) and target (Jetson Developer Kit) developer tools, APIs, and packages (OS images, tools, middleware, samples, and documentation) for developing on the NVIDIA Jetson Platform.

OS Images

Sample file systems derived from Ubuntu for Jetson Products.

Product L4T Version Notes

Jetson TX2

Jetson TX1

28.1

64-bit Ubuntu 16.04

Kernel 4.4

Production Release

Jetson TK1

21.5

Unchanged

32-bit Ubuntu 14.04

Kernel 3.10.40

Production Release













Libraries and APIs

  • CUDA 8.0 Toolkit for Host (Ubuntu with cross-development support)
  • CUDA 8.0 Toolkit for Jetson Device
  • cuDNN 6.0
  • TensorRT 2.1
  • VisionWorks 1.6
  • OpenCV4Tegra 2.4.13
  • MM API 28.1

Developer Tools

  • Tegra Graphics Debugger 2.4
    • A console-grade tool that enables debugging and profiling of applications using OpenGL ES 2.0-3.2 and OpenGL 4.2-4.5 core, plus many extensions, allowing developers to get the most out of the Jetson Platform.
  • Tegra System Profiler 3.7
    • A multi-core CPU sampling profiler that provides an interactive view of captured profiling data, helping improve overall application performance.

Samples

  • MM API Samples
  • VisionWorks Samples
  • CUDA Samples

Documentation

  • Jetson Documentation
  • JetPack Documentation

System Requirements

Host Platform Requirements:

  • Ubuntu Linux x64 (v14.04)
  • JetPack runs on the host Ubuntu x86_64 machine and sets up your developer environment and Jetson Developer Kit target via remote access
  • Note: A valid internet connection and at least 10GB of disk space is needed for the complete installation of JetPack.

Target Platform Requirements:

  • Jetson Developer Kit
    • USB Micro-B cable connecting Jetson to your Linux host for flashing
    • An HDMI cable connecting the HDMI port on Jetson Developer Kit to an external HDMI display (Not included in the developer kit)
    • An Ethernet cable plugged into the on-board Ethernet port, which is connected to either a secondary network card on your Linux host or the same network router prividing internet access for the Linux host (Not included in the developer kit)
    • To connect USB peripherals such as keyboard and mouse, a USB hub could be connected to the USB port on the Jetson system (Not included in the developer kit)

To access older versions of JetPack, please visit JetPack Archive.