AI is moving from the digital world into the physical one. Across factory floors and operating rooms, machines are evolving into collaborators that can see, sense and make decisions in real time.
To accelerate this transformation, NVIDIA today unveiled NVIDIA IGX Thor, a powerful, industrial-grade platform built to bring real-time physical AI directly to the edge, combining high-speed sensor processing, enterprise-grade reliability and functional safety in a small module for the desktop.
Delivering up to 8x the AI compute performance of its predecessor, NVIDIA IGX Orin, IGX Thor enables developers to build intelligent systems that perceive, reason and act faster, safer and smarter than ever.
Early adopters include industrial, robotic, medical and healthcare leaders, Diligent Robotics, EndoQuest Robotics, Hitachi Rail, Joby Aviation, Maven and SETI Institute, while CMR Surgical is evaluating IGX Thor to advance its medical capabilities.
IGX Thor Delivers Breakthrough Performance for Edge AI Applications
Medical and industrial companies have historically faced limitations advancing applications for medical imaging, surgical robotics and industrial AI automation due to edge systems lacking the speed and capacity to process real-time data or securely run multiple generative AI models without sacrificing performance.
IGX Thor overcomes these challenges by delivering robust, reliable AI compute tailored for industrial and medical environments. It provides real-time intelligence, seamless data connectivity, and built-in safety and security.
The platform features two types of NVIDIA Blackwell GPUs — an integrated GPU (iGPU) and a discrete GPU (dGPU) — delivering 5,581 FP4 teraflops of AI compute with 400 GbE connectivity. Compared with NVIDIA IGX Orin, IGX Thor provides up to 8x higher AI compute on iGPUs, 2.5x higher AI compute on dGPUs and 2x better connectivity to seamlessly run large language models and vision language models at the edge.
The industrial-grade platform comes with a 10-year lifecycle and long-term support for the NVIDIA AI software stack, keeping enterprise and physical AI applications accelerated and secured even in the most challenging conditions. IGX Thor runs the NVIDIA AI Enterprise software stack, including NVIDIA NIM microservices, which accelerate physical AI application development from the cloud to the edge, NVIDIA Isaac for robotics, NVIDIA Metropolis for visual AI and NVIDIA Holoscan for sensor processing.
In addition, NVIDIA IGX Thor integrates elements from the NVIDIA Halos full-stack safety system to embed functional safety into robotics, medical and industrial AI systems — tapping into onboard and infrastructure sensors to ensure both traditional and outside-in safety for safe human collaboration.
NVIDIA Partner Ecosystem Accelerates the Industrial and Medical Edge
Global industrial and robotic leaders Hitachi Rail, Maven, SETI Institute and Joby Aviation are adopting IGX Thor to build and deploy advanced AI solutions that enhance safety, improve efficiency and enable new levels of automation in factories, warehouses and transportation systems.
Hitachi Rail is using IGX Thor to deploy advanced predictive maintenance and autonomous inspection systems on rail networks, boosting operational efficiency and reliability.
“AI and data are transforming railways,” said Giuseppe Marino, group CEO of Hitachi Rail. “By adopting NVIDIA IGX Thor, we are bringing the world’s most powerful industrial-grade, real-time AI performance directly to the edge, enabling operators to better optimize their railways and infrastructure. This capability will strengthen reliability, efficiency and optimization for passengers and operators alike.”
Maven Robotics will integrate NVIDIA IGX Thor into its next-generation industrial robots to unite advanced embodied-AI performance with built-in safety and real-time decision-making at scale.
“Maven Robotics is bringing general-purpose robots to industrial environments, partnering with leading global manufacturing and logistics organizations to deploy at scale,” said Hamza Derbas, CEO of Maven Robotics. “NVIDIA Thor IGX sits at the core of our next-gen robots, pairing safety-rated compute with the performance to run advanced embodied-AI models so capability and compliance move forward together.”
Global medical device and healthcare pioneers Diligent Robotics and Endoquest Robotics are adopting IGX Thor. CMR Surgical is evaluating IGX Thor to power advanced AI capabilities within its surgical robotics systems, enabling real-time analysis and adaptive decision-making in the operating room. The platform’s safety, reliability and compute performance support CMR’s vision to deliver intelligent assistance that enhances surgical precision, improves efficiency and results in better patient outcomes.
“Precision and patient safety are at the heart of every procedure,” said Chris Fryer, chief technology officer of CMR Surgical. “With IGX Thor, we have the potential to deploy next-generation AI assistants and real-time surgical guidance that process high-fidelity data in real time, allowing us to simplify complex procedures and enable safer, more intelligent minimal-access surgery.”
NVIDIA’s partner ecosystem of equipment manufacturers, including Advantech, ADLINK, ASRock Rack, Barco, Curtiss-Wright, Dedicated Computing, EIZO Rugged Solutions, Inventec, NexCOBOT – a NEXCOM company, Onyx, WOLF Advanced Technology and YUAN, provides a range of IGX Thor-powered edge servers, customized carrier boards, design services, cameras and sensors, along with AI and system software to speed solution development.
The NVIDIA IGX Thor platform includes two production-ready systems — the NVIDIA IGX T5000 module and the NVIDIA IGX T7000 board kit — built for diverse physical AI applications with functional safety. The IGX Thor production systems and developer kits are expected to be available in December.
Watch the GTC Washington, D.C., keynote from NVIDIA founder and CEO Jensen Huang and explore sessions.