NVIDIA’s launch of the Blackwell-powered Jetson Thor signals a defining moment in robotics, delivering unmatched AI compute performance and efficiency for developers building the next generation of humanoid, industrial, and healthcare robots.
NVIDIA has officially announced the global availability of the Jetson AGX Thor developer kit and production modules, marking a major advancement in physical AI.
The system, powered by the Blackwell architecture, offers 7.5x more AI compute and 3.5x greater efficiency than its predecessor, Jetson Orin. This leap forward aims to accelerate adoption of robotics across industries from logistics to healthcare.
Next-Generation AI for Robotics
The Jetson Thor platform features an NVIDIA Blackwell GPU and 128GB of memory, delivering up to 2,070 FP4 teraflops of AI compute in a compact 130-watt envelope. This performance makes it possible to run advanced generative AI models at the edge, including large language and vision language models.
Compared to Jetson Orin, Jetson Thor brings transformative gains in compute and energy efficiency. These enhancements are essential for real-time inference, a requirement for humanoid robotics, surgical assistance, and automated systems that need split-second reasoning in dynamic environments.
Industry Adoption and Applications
Global leaders in robotics and automation are already integrating Jetson Thor into their systems. Early adopters include Agility Robotics, Amazon Robotics, Boston Dynamics, Caterpillar, Figure, Hexagon, Medtronic, and Meta. Each company is exploring how Jetson Thor’s AI capacity can support unique industry challenges.
Other organizations such as John Deere, OpenAI, and Physical Intelligence are currently evaluating the platform. Their focus ranges from precision agriculture to AI-driven humanoids capable of interacting seamlessly with people and the physical world.
Real-World Use Cases
- Manufacturing & Logistics: Enhanced decision-making for autonomous robots managing warehouses and global supply chains.
- Healthcare & Surgery: Enabling surgical robots to process real-time data and assist physicians with greater accuracy and safety.
Solving Core Robotics Challenges
Robotics systems require more than raw compute power. They must operate under strict energy limits while supporting multiple AI workflows simultaneously. Jetson Thor addresses these challenges with its compact design and ability to run generative AI at the edge.
“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, NVIDIA’s founder and CEO.
“With unmatched performance and energy efficiency, Jetson Thor is the ultimate supercomputer to drive the age of physical AI and general robotics.”
Comparative Performance
The following performance comparison highlights Jetson Thor’s leap over Jetson Orin:
| Feature | Jetson Orin | Jetson Thor |
|---|---|---|
| AI Compute | ~275 FP4 teraflops | 2,070 FP4 teraflops |
| Energy Efficiency | Baseline | 3.5x greater |
| Generative AI Capability | Limited | Supports multi-model inference at edge |
Voices From Industry Leaders
Brett Adcock, CEO of Figure, emphasized Jetson Thor’s impact on humanoids:
“NVIDIA Jetson Thor’s server-class performance, delivered within a compact and power-efficient design, allows us to deploy the large-scale generative AI models necessary for our humanoids to perceive, reason and act in complex environments.”
Amazon Robotics also praised its potential.
“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,”
said Tye Brady, chief technologist at Amazon Robotics.
Caterpillar added perspective from industrial operations.
“NVIDIA Jetson Thor offers the AI performance we need to develop and deploy the construction and mining equipment of the future, enhancing precision, reducing waste and improving safety,”
said Joe Creed, CEO of Caterpillar.
Availability and Pricing
The NVIDIA Jetson AGX Thor developer kit is available now starting at $3,499, while Jetson T5000 production modules can be sourced through worldwide distribution partners.
Carrier boards and production systems are also available through embedded hardware partners, expanding accessibility for developers worldwide.
The Takeaway
NVIDIA’s Jetson Thor marks a pivotal step toward realizing the promise of general robotics. With its Blackwell-powered architecture, significant performance boost, and industry adoption, it positions itself as the cornerstone for next-generation autonomous systems.
As more industries adopt physical AI, Jetson Thor’s influence is set to grow in shaping how humans and machines collaborate in real time.
Sources: NVIDIA.
Prepared by Ivan Alexander Golden, Founder of THX News™, an independent news organization delivering timely insights from global official sources. Combines AI-analyzed research with human-edited accuracy and context.
