NVIDIA’s new Jetson Thor modules are set to revolutionize robotics by providing a significant boost in AI compute power, CPU performance, and memory. The modules, now in general availability, offer 7.5x more AI compute, 3.1x more CPU performance, and 2x more memory than their predecessor, the NVIDIA Jetson Orin.
These advancements enable robots to process high-speed sensor data and perform visual reasoning at the edge, opening up new possibilities for multimodal AI applications like humanoid robotics. Agility Robotics, a leader in humanoid robotics, has integrated Jetson Thor into its fifth-generation robot, Digit, to enhance real-time perception and decision-making capabilities.
Boston Dynamics is also integrating Jetson Thor into its humanoid robot Atlas, enabling it to harness formerly server-level compute, AI workload acceleration, high-bandwidth data processing, and significant memory on device. The modules will accelerate various robotic applications, including surgical assistants, smart tractors, delivery robots, industrial manipulators, and visual AI agents.
Jetson Thor is built for generative reasoning models and enables the next generation of physical AI agents to run in real-time at the edge while minimizing cloud dependency. It supports popular generative AI frameworks and AI reasoning models with unmatched real-time performance.
The modules also run the full NVIDIA AI software stack, accelerating virtually every physical AI workflow with platforms like NVIDIA Isaac for robotics, NVIDIA Metropolis for video analytics AI agents, and NVIDIA Holoscan for sensor processing.
Research labs at Stanford University, Carnegie Mellon University, and the University of Zurich are tapping Jetson Thor to push the boundaries of perception, planning, and navigation models for various applications. The NVIDIA Jetson AGX Thor developer kit is available now starting at $3,499, while the Jetson T5000 modules start at $2,999.
Source: https://blogs.nvidia.com/blog/jetson-thor-physical-ai-edge