NVIDIA has officially launched the Jetson AGX Thor developer kit and production module, alongside a powerful robotics computer designed to empower millions of robots across industries such as manufacturing, logistics, transportation, healthcare, agriculture, and retail. Early adopters of this cutting-edge technology include AgiBot, Engine AI, Galaxy General, UBTECH, United Imaging, Yushu, and Vanjee.
“We built Jetson Thor for the millions of developers working on robotic systems that will interact with and increasingly shape the physical world,” said Jensen Huang, founder and CEO of NVIDIA. “With its exceptional performance and power efficiency, and its ability to run multiple generative AI models at the edge, Jetson Thor is the ultimate supercomputer to power the era of physical AI and general-purpose robotics.”
AI Authority Trend: Lambda Expands AI Factories with Supermicro NVIDIA Blackwell Servers
With NVIDIA Blackwell GPUs and 128GB of video memory, Jetson Thor packs an impressive punch delivering up to 2,070 FP4 teraflops of AI computing power while keeping energy use under 130 watts. This represents up to 7.5 times the AI computing power and 3.5 times the energy efficiency of its predecessor, the Jetson AGX Orin, making it capable of running advanced AI models, including vision-language-behavior (VLA) models like the NVIDIA Isaac GR00T N1.5, as well as popular large language and vision-language models.
Moreover, Jetson Thor addresses one of the key challenges in robotics development: running multiple AI workflows for real-time, intelligent interaction between robots, humans, and the physical world. By enabling real-time inference, the platform powers high-performance AI in action whether it’s humanoid robots, automated farming, or even assisting in surgery.
Leading robotics companies around the world are already turning to Jetson Thor for their development projects. The system integrates seamlessly with the full-stack NVIDIA Jetson software platform, supporting all popular AI frameworks and generative AI models. It works seamlessly with NVIDIA’s cloud-to-edge software stack, including Isaac for robotics simulation, the Isaac GR00T humanoid robot base, Metropolis for vision AI, and Holoscan for real-time processing.
AI Authority Trend: NVIDIA Unveils Spectrum-XGS Ethernet to Power Giga-Scale AI Super-Factories
“At Yushu, our goal is to push the boundaries of what’s possible with high-performance, dynamic robots, providing friendly, safe robots for everyone,” said Xingxing Wang, founder and CEO of Yushu. “Jetson Thor delivers a quantum leap in computing power, enhancing agility, accelerating decision-making, and bringing a whole new level of autonomy.”
Similarly, Wang He, founder and CTO of Galaxy General, added, “Our G1 Premium has already seen significant improvements in speed and fluidity using Jetson Thor, and we’re eager to unlock even more powerful capabilities in large VLA models.”
The NVIDIA Jetson AGX Thor Developer Kit is now available starting at $3,499. The Jetson T5000 production module can be purchased globally through distribution partners, along production systems and carrier boards offered by embedded partners.
AI Authority Trend: DataPelago Nucleus Outperforms Nvidia cuDF in New Benchmark Results
To share your insights, please write to us at sudipto@intentamplify.com
