Nvidia CEO Unveils Powerful New ‘Brain’ for Humanoid Robots

Nvidia CEO Unveils Powerful New 'Brain' for Humanoid Robots

In an exciting advancement for the field of robotics, Nvidia has unveiled the Jetson Thor, a new module designed to significantly enhance the capabilities of robots worldwide. With this launch, robotics engineers are set to experience a surge in innovation.

Nvidia introduced Thor through a captivating video where CEO Jensen Huang presents a humanoid robot with a gift box labeled “Enjoy your new brain.” As the box opens, multiple robots reveal the advanced device aimed at making them smarter and more efficient.

Unmatched AI Computing Power

The Jetson Thor module is engineered for various robotic applications and claims an astounding 7.5 times more AI computing power compared to its predecessor, Orin, which was released earlier this year. Additionally, it boasts 3.1 times more CPU performance and features double the memory capacity.

Real-Time Robotics and Enhanced Performance

Thor is specifically built for real-time robotics tasks, adeptly managing high-bandwidth multi-sensor inputs with ultra-low latency. Its architecture allows it to execute large generative AI and transformer-based models locally, minimizing dependency on cloud infrastructure.

Industry Applications: Real-World Impact

Prominent robotics firms are already integrating the Thor module into their systems. For instance, Boston Dynamics is utilizing Thor in its Atlas humanoid robot, enabling it to process complex multimodal AI models in real-time, thus improving its perception, decision-making, and body manipulation abilities.

Similarly, Agility Robotics is incorporating Thor into its sixth-generation Digit humanoid robot to enhance its perception and adaptability in warehouse environments. According to Peggy Johnson, CEO of Agility Robotics, “The powerful edge processing offered by Jetson Thor will take Digit to the next level, enhancing its real-time responsiveness and expanding its capabilities in more complex scenarios.”

Educational Institutions Leading the Way

Thor is not only attracting big-name companies but also garnering attention from leading academic institutions such as Carnegie Mellon, Stanford, and the University of Zurich. These tech innovators will leverage Thor for advanced perception, navigation, robot fleets, and sensor fusion projects.

Widespread Utilization of Jetson Chips

This latest technology is already being utilized by major players, including Amazon, Meta, Figure, and Hexagon, with reports indicating that companies like OpenAI are exploring the potential of these chips for their solutions.

Future Growth in Robotics

Currently, robotics accounts for only 1% of Nvidia’s revenue. However, Huang highlights this sector as a significant growth opportunity moving forward, given the rapid technological advancements in AI and machine learning.

Nvidia has announced the Jetson AGX Thor developer kit, now available starting at $3,499 (approximately €3,340). For bulk purchases, Nvidia Jetson T5000 modules begin at $2,999 (around €2,742) for 1,000 units, available through authorized partners.

Additionally, Nvidia introduced the DRIVE AGX Thor developer kit focused on autonomous vehicles, which is available for preorder with initial shipments expected in September 2025.

How does the Jetson Thor enhance robotics performance?

The Jetson Thor enhances robotics by providing superior AI computing power, enabling real-time processing of complex data, thus improving responsiveness and capabilities in various environments.

What types of robots can use the Jetson Thor module?

The Jetson Thor module is versatile and can be used in various types of robots, including humanoid robots like Atlas and more specialized models such as Agility Robotics’ Digit.

Why is Nvidia focused on robotics as a growth sector?

Nvidia sees robotics as a key growth opportunity because it currently represents a small percentage of its overall revenue yet holds significant potential for advancements in AI and machine learning technologies.

Are other companies using Nvidia’s Jetson chips?

Yes, a number of high-profile companies, including Amazon and Meta, are utilizing Nvidia’s Jetson chips, with many others exploring possibilities for integration into their operations.

With the dynamic advancements in robotics and AI technology, there’s never been a better time to stay informed. Continue exploring related content on this topic at Moyens I/O.