Product Description - Jetson Xavier NX

NVIDIA® Jetson Xavier™ NX is a small system-on-module (SOM) that delivers supercomputer performance. With accelerated computing of up to 21 trillion operations per second (TOPS), it provides sufficient power for users to run multiple modern neural networks in parallel and process data from multiple high-resolution sensors, meeting the requirements of a complete AI system. The Jetson Xavier NX is ready for production and supports all common AI frameworks.

Now Cloud-Native Ready

The Jetson Xavier NX module now supports cloud-native capabilities, enabling developers to create and deploy high-quality, software-defined features on embedded and edge devices. With pre-trained AI models from NVIDIA NGC and the TAO Toolkit, AI networks can be trained and optimized faster and deployed to Jetson devices using containerized solutions, ensuring flexible and smooth updates. The performance of Jetson Xavier NX is 10 times that of the widely adopted previous generation, Jetson TX2, when accelerating the NVIDIA software stack.

Small Size, Powerful Edge Performance

Performance of Xavier, Size of Nano

The Jetson Xavier NX, measuring only 70 mm x 45 mm, packs the capabilities of the NVIDIA Xavier SoC into a module the size of Jetson Nano™. This powerful, compact module combines outstanding performance and energy efficiency, featuring a rich array of I/O options from high-speed CSI and PCIe to low-speed I2C and GPIO. With its small size, multiple sensor interfaces, and powerful performance, it brings new capabilities to your embedded AI and edge systems.

Incredible Energy Efficiency

The Jetson Xavier NX supports multiple power modes, including low-power modes suitable for battery-operated systems. It can deliver up to 14 TOPS of compute performance at as low as 10W power consumption. This allows more power budget to be reserved for sensors and peripheral devices while fully utilizing the NVIDIA software stack. With sufficient performance to run all modern AI networks and frameworks, it leverages deep learning, computer vision, computer graphics, and multimedia along with other accelerated libraries.

Suitable for Critical Embedded Applications

Jetson Xavier NX is ideal for high-performance AI systems such as commercial robots, medical instruments, smart cameras, high-resolution sensors, automated optical inspection, smart factories, and other AIoT embedded systems. This module is energy-efficient, smaller than a credit card (70x45mm), and offers server-class performance with 21 TOPS (at 15W or 20W) or up to 14 TOPS (at 10W). It can run multiple modern neural networks simultaneously and process data from multiple high-resolution sensors, meeting the needs of a complete AI system. With cloud-native support, AI software development and deployment on edge devices become more manageable.

Jetson Xavier NX is ready for mass production and supports all common AI frameworks, opening new opportunities for embedded edge computing devices to handle increased performance demands from AI workloads within limited size, weight, power budget, or cost constraints.

Jetson Xavier NX Module Technical Specifications

Specification	Details
AI Performance	21 TOPS
GPU	384-core NVIDIA Volta™ GPU with 48 Tensor Cores
CPU	6-core NVIDIA Carmel ARM® v8.2 64-bit CPU, 6MB L2 + 4MB L3
Memory	8 GB 128-bit LPDDR4x, 59.7 GB/s
Storage	16 GB eMMC 5.1
Power Consumption	10W, 15W, 20W
PCIe	1 x1 (PCIe Gen3) + 1 x4 (PCIe Gen4), total 144 GT/s
CSI Camera	Up to 6 cameras (24 via virtual channels), 14 lanes (3x4 or 6x2 or 3x4 + 1x2 or 5x2 + 1x4) MIPI CSI-2, D-PHY 1.2 (up to 30 Gbps)
Video Encode	2x 4K60, 4x 4K30, 10x 1080p60, 22x 1080p30 (H.265) 2x 4K60, 4x 4K30, 10x 1080p60, 20x 1080p30 (H.264)
Video Decode	2x 8K30, 6x 4K60, 12x 4K30, 22x 1080p60, 44x 1080p30 (H.265) 2x 4K60, 6x 4K30, 10x 1080p60, 22x 1080p30 (H.264)
Display	2 multi-mode DP 1.4/eDP 1.4/HDMI 2.0
DL Accelerator	2x NVDLA
Vision Accelerator	2x PVA
Networking	10/100/1000 BASE-T Ethernet
Mechanical	69.6mm x 45mm, 260-pin SO-DIMM connector

* Please refer to the "Software Features" section in the latest NVIDIA Jetson Linux Developer Guide for the full list of supported features.