- Nvidia Jetson Thor is framed as the robot brain for physical AI
- Major robotics companies are already listed as early adopters worldwide
- Nvidia Jetson Thor offers 2,070 FP4 teraflops within a 130-watt power envelope
Nvidia has released the Jetson AGX Thor developer kit, calling it the next step toward robotics systems which can function in real time.
The system, built on the Blackwell GPU line, is framed as a platform for “physical AI” and advanced robotic functions across manufacturing, logistics, healthcare, farming, retail, and transport.
Nvidia says it can deliver up to 7.5 times more AI compute and over three times the energy efficiency of its Jetson Orin line, which has been in wide use since 2022.
Offers supercomputer-level capacity
Nvidia went on to describe Jetson Thor as “the ultimate supercomputer to drive the age of physical AI and general robotics.”
“We’ve built Jetson Thor for the millions of developers working on robotic systems that interact with and increasingly shape the physical world,” said Jensen Huang, founder and CEO of Nvidia.
“With unmatched performance and energy efficiency, and the ability to run multiple generative AI models at the edge, Jetson Thor is the ultimate supercomputer to drive the age of physical AI and general robotics.”
With a quoted figure of 2,070 FP4 teraflops in a 130-watt envelope, it is positioned as powerful enough to run multiple generative models at once.
It supports vision-language-action models like Isaac GR00T N1.5, along with other LLM systems.
The device also integrates 128GB of memory, which is expected to make it capable of handling larger AI workflows at the edge.
Several robotics players are already listed as early adopters, including Agility Robotics, Amazon Robotics, Boston Dynamics, Caterpillar, Hexagon, and Medtronic.
Meta has also been named as an early partner, while companies such as John Deere, OpenAI, and Physical Intelligence are said to be testing the system.
“Nvidia Jetson Thor offers the computational horsepower and energy efficiency necessary to develop and scale the next generation of AI-powered robots that can operate safely and effectively in dynamic, real-world environments, transforming how we move and manage goods globally,” said Tye Brady, chief technologist at Amazon Robotics.
Nvidia notes more than two million developers already use its robotics stack, with over 7,000 customers having deployed Jetson Orin hardware in edge AI projects.
Jetson Thor runs on the Nvidia Jetson software platform, which is designed to support multiple AI tools at once.
The package integrates with Nvidia Isaac for simulation, Metropolis for vision AI, and Holoscan for real-time sensor processing.
This arrangement is intended to allow one system-on-module to support many AI writer models and workflows, rather than requiring several separate chips.
The developer kit is available now at $3,499 and the production systems, including carrier boards, will be distributed worldwide through its partners.
