Sensor data is like brain food for robots, and as robots become more complex and sophisticated in their ability to take in many more data streams for many more sensors they will need more powerful AI brains to process all of that data.
Nvidia’s AGX Thor edge computing module, the successor to the company’s Jetson Orin and built on the newer Blackwell GPU, is here to embrace that challenge. Nvidia announced general availability for its AGX Thor developer kit and production modules, a move that comes almost a year after Nvidia announced AGX Thor.
The release should provide an adrenaline boost to the robotics projects of AGX Thor early adopters, such as Agility Robotics, Amazon Robotics, Boston Dynamics, Caterpillar, Figure, Hexagon, Medtronic and Meta, and the more than 2 million robotics developers using Nvidia’s robotics stack. Among the early adopters, Nvidia partner Hexagon back in June showcased its latest humanoid robot, AEON.
“[Jetson AGX Thor is] the physical computer that's inside the robot, the so-called robot brain,” said Deepu Talla, in a media briefing. He later added, “It is the ultimate platform for physical AI and robotics, built for the age of reasoning AI running at the edge, low power, high performance, real time.”
Thor provides 7.5x more AI compute and 3.5x greater energy efficiency compared with Jetson Orin, Nvidia said. The Blackwell GPU provides the foundation for 128GB of memory and up to 2,070 FP4 teraflops of AI compute with just a 130-watt power envelope. It is evidence of how Nvidia is pushing Blackwell to all corners of its ecosystem, from massive data centers focused on AI training to edge computing systems more focused on inference.
Agility Robotic is planning to use Jetson AGX Thor as the on-board compute platform for the sixth generation of its Digit robot–a commercially-deployed robot working in warehouse and manufacturing environments–to enhance Digit’s real-time perception and decision-making capabilities, supporting increasingly complex AI skills and behaviors, according to Nvidia.
“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,” said Peggy Johnson, CEO of Agility Robotics. “With Jetson Thor, we can deliver the latest physical AI advancements to optimize operations across our customers’ warehouses and factories.”
General availability for Jetson AGX Thor comes just a couple of days before Nvidia is scheduled to report its latest quarterly earnings. As of the company’s fiscal first quarter earlier this year, revenue from its Automotive and Robotics business accounted for just 2% of overall company revenue, but Nvidia CEO Jensen Huang has increased his focus on robotics in public speeches in recent months, often talking about a vision for “a billion robots around the world.”