AI Business is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 3099067.

Nvidia's Jetson TX2 Offers Big AI Power in a Small Size

by Ed Lauder
Article Image

Nvidia's new Jetson TX2 supports artificial intelligence computing on a system the size of a credit card.

Computing companies are always looking for new ways get more power in a smaller form factor. Nvidia is one of the leaders in that area. Their computer chips are some of the smallest and most powerful on the market and their Tigra processors operate most of our mobile devices: tables, mobile phones and even some mobile gaming platforms, including Nintendo's most recent home/mobile console hybrid, the Nintendo Switch.

The company's now announced at a launch event in San Fransisco, via techradar, the Jetson TX2, which is the next step in AI computing. It will support AI in factory robots, drones and even smart video cameras scattered around our cities. The TX2 offers all of this AI computing power in a package the size of a standard credit card, 50mm x 87mm to be exact.

Nvidia's previous AI computer chips were the Jetson TX1 and TK1, yet the company claims the the TX2 has twice as much performance and efficiency as their old chips, and its functions will include voice and facial recognition, navigation and object detection. Perfect for drones and self-driving cars, which are areas that Nvidia is very keen on developing, especially self-driving cars.

During the TX2's launch event in San Fransisco, Nvidia demonstrated how it could recognise other vehicles on the road, even in rainy conditions. Nvidia developed the Jetson TK2 in partnership with Cisco, Live Planet, MIT, Toyota and Toshiba.

Packed inside its tiny platform the Jetson TX2 has a 256-core Nvidia Pascal-based GPU, a CPU comprised of a dual 64-bit Nvidia Denver 2 chips and a Quad ARM A57, and 8GB of LPDDR4 RAM. It also features 32GB of storage, Bluetooth connectivity, and supports 4K video. That's a lot of tech to fit into such a small package.

Image courtesy of BagoGames

Practitioner Portal - for AI practitioners


Facebook launches Dynabench to test AI in realistic conditions


Looking to solve the problem of users messing with its algorithms


InterDigital releases AI-based video compression codec tool


Available on GitHub under Apache 2.0 license

Practitioner Portal


More EBooks

Upcoming Webinars

More Webinars
AI Knowledge Hub

Experts in AI

Partner Perspectives

content from our sponsors

Research Reports

More Research Reports


Smart Building AI

Infographics archive

Newsletter Sign Up

Sign Up