Select Page

Advance Innovation Today


Hadron delivers high performance deep learning-based imaging.


An out-of-the-box AI Solution

Hadron is a compact yet powerful embedded device vision device designed for industrial automation and autonomous robots. Powered by the NVIDIA® Jetson TX2, Hadron delivers best-in-class performance and efficiency. Dual 4K Sony camera sensors provide high-fidelity video and Simultaneous Localization and Mapping (SLAM) capability. Combined with specialized low-latency drivers, Hadron offers performance and functionality that is unrivaled. Hadron comes preinstalled with the most commonly used open-source tools including Tensorflow, OpenCV, ROS, and can be configured to perfectly fit your application.


Focus on your research, not your equipment. We provide all the tools you need to get to work quickly. Hadron includes all the hardware and software you need to develop your robotic or automation application.


Building the systems to run your projects is both time-consuming and expensive. Hadron is an integrated solution providing hardware designed for your workload, without costing you a premium.

High Performance

Hadron utilizes the only embedded processor built specifically for AI workloads. Combined with an ultra-low latency stereo camera, you can minimize network bandwidth via on-board image processing and transmit just the relevant data.


Hadron is modular and upgradable, supporting a variety of accessories that further expand functionality. It comes with all the I/O you expect: dual USB 3.0, USB OTG, Debug, CAN (DeviceNet), Gigabit Ethernet, WiFi, and Bluetooth are all standard.

© 2018 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, Pascal, and DIGITS are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. All other trademarks and copyrights are the property of their respective owners.

Top Performance and Efficiency

Hadron utilizes the NVIDIA® Jetson® TX2 embedded processor to deliver the performance to run the most demanding deep learning and computer vision applications with minimal power consumption. 

The Jetson TX2 is the fastest, most power-efficient embedded AI computing device, built around the NVIDIA Pascal GPU architecture and containing 256 CUDA cores. 

Hadron comes with all Jetson Development Pack (Jetpack) tools pre-installed:

Deep Learning: TensorRT, cuDNN, NVIDIA DIGITS Workflow

Computer Vision: NVIDIA VisionWorks, OpenCV

GPU Compute: NVIDIA CUDA, CUDA libraries

Robotics: ROS

High-Resolution Stereo Vision

Hadron comes with high-resolution stereo vision standard. Dual Sony IMX377 sensors deliver stunning 4K image quality over a high-speed CSI interface. Custom made drivers deliver video with very low latency and enable a variety of configuration options to meet your specific requirements. The cameras support a variety of lens mount options including M12, D14, and CS lens mounts. Varioptics Liquid Lenses are available, should you require high speed auto-focus, as is thermal vision with optional FLIR LWIR cameras.



Deep Learning

Industrial Automation

Stereo Vision

Autonomous Vehicles

GPU NVIDIA Pascal 256 CUDA cores
CPU HMP Dual Denver 2/2 MB L2 + Quad ARM® A57/2 MB L2
Video Support 4K x 2K 60 Hz Encode (HEVC), 4K x 2K 60 Hz Decode (12-Bit Support)
Memory 8 GB 128 bit LPDDR4, 59.7 GB/s
Display Video out
Storage 32 GB eMMC, microSD expandable
Cameras Dual Sony IMX377 4K CMOS sensors
Connectivity Gigabit Ethernet, 802.11ac WLAN, Bluetooth
I/O 2x USB 3.0, Debug, USB OTG, 2x CAN (DeviceNet)
Lens Support CS (stock), M12, D14
Optional Accessories Varioptics Liquid Lens, Thermal Vision (FLIR LWIR), Customizable electronic controlled band pass/cut filter, Universal Camera Mount (1/4″ UNC), Arca-Swiss Quick Release Mount