Nvidia Debuts Super Advanced Jetson Orin Nano System For AI And Robots

JAKARTA - Nvidia has just introduced the Jetson Orin Nano, a system-on-module (SOM) that will support the next generation of entry-level AI and robotics. This system is planned to be launched as soon as next year.

The SOM itself is board based and may have room to include additional components, it could even include an SOC. In short, SOM is a ready-to-use computing solution, but not a full-fledged computer.

During yesterday's GTC 2022 keynote, Nvidia said the new version provides an 80x performance boost over the first-generation Jetson Nano.

The original version was released in 2019 and has been used as an early entry into the world of AI and robotics, especially for STEM enthusiasts and students. It is claimed, this new version looks serious in terms of increasing power.

The Jetson Orin Nano delivers up to 40 trillion operations per second (TOPS), which is the number of compute operations the chip can handle at 100 percent utilization in the smallest Jetson form factor to date.

The Jetson Orin Nano is currently in the entry-level lineup of the Jetson family, including six Orin-based production modules aimed at a variety of robotics and local offline computing applications.

Coming in a module compatible with Nvidia's previously announced Jetson Orin NX, the Jetson Orin Nano supports AI application channels with Ampere architecture GPUs. Ampere itself is the GPU architecture that Nvidia launched in 2020.

Launching Digital Trends, Wednesday, September 21, all versions will be available in January starting with the Jetson Orin Nano with a capacity of 8GB, which provides up to 40 TOPS with power that can be configured from 7W to 15W, and while the Jetson Orin Nano with a capacity of 4GB can reach up to 20 TOPS with power options as low as 5W to 10W.

"Over 1,000 customers and 150 partners have used the Jetson AGX Orin since Nvidia announced its availability just six months ago, and the Orin Nano will significantly expand this adoption," said Deepu Talla, VP of Embedded and edge computing at Nvidia.

"With tremendous performance improvements for millions of AI and (robotic) developers, the Jetson Orin Nano is setting a new standard for entry-level AI and robotics."