- Nvidia jetson thor is framed as robot brain to physical AI
- Larger robotics companies are already listed as early adoptors all over the world
- NVIDIA Jetson Thor offers 2,070 FP4 Teraflops within a 130-watt power volly
Nvidia has released Jetson Agx Thor Developer Kit and calls it the next step against robotics systems that can work in real time.
The system, built on the Blackwell GPU line, is framed as a platform for “Physical AI” and advanced robotic functions across manufacture, logistics, healthcare, agriculture, retail and transport.
Nvidia says it can deliver up to 7.5 times more AI calculation and over three times the energy efficiency of its Jetson Orin line, which has been in broad use since 2022.
Offers capacity at supercomputer level
Nvidia continued to describe Jetson Thor as “the ultimate supercomputer to drive the age of physical AI and general robotics.”
“We have built Jetson Thor for the millions of developers working on robot 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 several 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 concentrate, it is placed as powerful enough to run several generative models at once.
It supports models with vision-language action such as ISAAC GR00T N1.5 along with other LLM systems.
The device also integrates 128 GB of memory, which is expected to enable it to handle larger AI workflows at the edge.
Several robotic 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 like John Deere, Openai and physical intelligence are said to test the system.
“Nvidia Jetson Thor offers the computational horsepower and energy efficiency needed to develop and scale the next generation of AI-driven robots that can work safely and efficiently in dynamic environments in the real world and transform how we move and control goods globally,” said Tye Brady, chief technologist at Amazon Robotics.
Nvidia notes that more than two million developers are already using its robotic table, where over 7,000 customers have implemented Jetson Orin hardware in Kant AI projects.
Jetson Thor runs on the Nvidia Jetson Software Platform, which is designed to support multiple AI tools at once.
The package is integrated with Nvidia Isaac for simulation, Metropolis for Vision AI and Holoscan for real -time sensor treatment.
This scheme is intended to allow a system-on-module to support many AI author models and workflows instead of requiring more separate chips.
The developer set is now available for $ 3,499, and the production systems, including transport boards, will be distributed worldwide through its partners.
