NVIDIA announced at COMPUTEX the release of JetPack 7.2 and NemoClaw support for its Jetson platform, marking a significant step in bringing agentic AI into physical-world applications. The update includes enhancements to the Jetson software stack, such as Yocto project support, CUDA 13 on Jetson Orin, and a performance boost for the Jetson AGX Orin 32GB module, reaching 241 TOPS of AI compute. The release also introduces Multi-Instance GPU (MIG) support on NVIDIA Jetson Thor, allowing developers to reserve dedicated GPU resources for deterministic workloads. 'Agentic AI is here, and Jetson’s programmability and high performance enable developers to instantly deploy physical AI agents in production at the edge,' said Deepu Talla, vice president of robotics and edge computing at NVIDIA. The JetPack 7.2 release includes three layers: a base layer with operating system and compute support, a middle layer with agent skills for automating developer tasks, and NemoClaw at the top for deploying agentic AI on Jetson. The new features aim to accelerate development and reduce time to market for edge AI applications. *Source: [nvidia](https://blogs.nvidia.com/blog/jetson-agentic-ai-physical-world/)*