NVIDIA's new Jetson Thor is here, called 'ChatGPT moment' for physical AI, robotics, smart EVs

NVIDIA's impressive new Jetson Thor system is now out in the market, a tiny compact powerhouse AI system, that Boston Dynamics is integrating into its humanoid robot Atlas.

3

NVIDIA has published a new blog post on its website detailing the new Jetson Thor chip, which features 7.5x more AI compute power, 3.1x more CPU performance, and 2x more memory than its predecessor, the Jetson Orin. This huge performance leap enables roboticists to process high-speed sensor data and perform visual reasoning at the edge, these are workflows that were too slow to run in dynamic real-world environments.

NVIDIA's new Jetson Thor chip and its new powerful performance upgrades across the board open new possibilities for multimodal AI applications, such as humanoid robots. Jetson Thor modules will be used inside of new robotics computers that can "serve as the brains for robotic systems across research and industry", explains NVIDIA.

This has led to Boston Dynamics, which has been at the forefront of the industry's most advanced robots for over 30 years, using NVIDIA's new Jetson Thor inside of its humanoid robot Atlas, giving it server-level compute power, AI workload acceleration, data processing, and "significant" memory on-device.

Boston Dynamics will also be using Jetson Thor in other markets other than humanoid robots, using NVIDIA's new system for various robotic uses like surgical assistants, smart tractors, delivery robots, industrial manipulations, and visual AI agents. Each of them would feature real-time inference on-device allowing for larger, more complex AI models.

Agility Robotics is another leader in the humanoid robot business, has integrated NVIDIA Jetson into its new fifth-generation robot, Digit, with plans to use Jetson Thor as the onboard compute platform for its sixth-generation Digit. This move to the new NVIDIA Jetson Thor chip enhances Digit's real-time perception and decision-making capabilities, supporting increasingly complex AI skills and behaviors.

Digit is already commercially deployed with the humanoid robot performing logistics tasks like stacking, loading, and palletizing in warehouses and manufacturing facilities.

3

Peggy Johnson, CEO of Agility Robotics, said: "The powerful edge processing offered by Jetson Thor will take Digit to the next level - enhancing its real-time responsiveness and expanding its abilities to a broader, more complex set of skills. With Jetson Thor, we can deliver the latest physical AI advancements to optimize operations across our customers' warehouses and factories".