NVIDIA Jetson Brings Agentic AI to the Physical World

2026-06-02 · Source: NVIDIA Blog · Field: Technology & Digital — Robotics & Autonomous Systems, Artificial Intelligence & Machine Learning, Internet of Things (IoT) & Connected Devices · Depth: Advanced, short

Summary

NVIDIA announced JetPack 7.2 and NemoClaw support for its Jetson embedded systems at COMPUTEX, extending agentic AI capabilities to physical world applications like robotics and industrial automation. JetPack 7.2 introduces agentic AI skills, Yocto project support, NVIDIA CUDA 13 on Jetson Orin, and Multi-Instance GPU (MIG) support on Jetson Thor. It also boosts Jetson AGX Orin 32GB performance by 20% to 241 TOPS of AI compute. NemoClaw, NVIDIA's agentic AI framework, now deploys on the production-grade Jetson stack, enabling task automation for industrial systems. This release features three layers: JetPack 7.2 for OS and compute, a new agent skills layer for automating developer tasks, and NemoClaw for deploying agentic AI. Real-world deployments include Solomon's humanoid robots, Advantech's agentic factory brain, and SandStar's AI vending machines, which achieved nearly 40% memory optimization. NoTraffic also reported a 29% memory reduction in its traffic management systems.

Key takeaway

For MLOps Engineers deploying AI at the edge, NVIDIA's JetPack 7.2 and NemoClaw on Jetson significantly streamline physical AI agent development. You can now utilize production-grade agentic AI for robotics and industrial automation, reducing development time from weeks to days. Consider implementing Yocto-based OS support for memory-optimized deployments and Multi-Instance GPU for deterministic workloads to enhance reliability and efficiency. This update enables faster time-to-market and lower total cost of ownership for your edge AI solutions.

Key insights

NVIDIA Jetson now supports agentic AI via JetPack 7.2 and NemoClaw, enabling physical AI agents at the edge.

Principles

• Edge AI benefits from memory optimization.
• Deterministic workloads require dedicated GPU resources.
• Customizable OS improves industrial deployments.

Method

The release layers JetPack 7.2 (OS/compute), agent skills (automating developer tasks), and NemoClaw (agentic AI deployment) to accelerate edge AI system development.

In practice

• Deploy NemoClaw for industrial task automation.
• Use Yocto-based OS for memory-bound systems.
• Reserve GPU resources for real-time perception.

Topics

• Agentic AI
• NVIDIA Jetson
• Edge AI
• Robotics
• Industrial Automation
• Yocto Project
• NemoClaw

Code references

• NVIDIA-AI-Blueprints/video-search-and-summarization

Best for: AI Architect, Machine Learning Engineer, CTO, Robotics Engineer, AI Engineer, MLOps Engineer

Related on AIssential

• Last week in AI | 23 March — NVIDIA is integrating hardware and software into "AI factory" platforms, while new research enhances LLM multi-step task capabilities.
• Deploying Open Source Vision Language Models (VLM) on Jetson — Deploying VLMs like Cosmos Reason 2B on NVIDIA Jetson devices enables real-time, interactive physical AI at the edge.
• MiniMax M2.7 Advances Scalable Agentic Workflows on NVIDIA Platforms for Complex AI Applications — MiniMax M2.7, a 230B-parameter MoE model, offers high efficiency and performance for agentic tasks via NVIDIA optimizations.
• Build enterprise-ready Agentic AI with DataRobot using NVIDIA Nemotron 3 Super — Transforming powerful LLMs into trusted, production-grade enterprise AI systems requires robust infrastructure beyond model capabilities.
• Nvidia’s version of OpenClaw could solve its biggest problem: security — Nvidia's NemoClaw offers an enterprise-grade, secure platform for deploying and managing AI agents based on the OpenClaw framework.
• As Open Models Spark AI Boom, NVIDIA Jetson Brings It to Life at the Edge — Edge AI platforms like NVIDIA Jetson enable real-time, private, and cost-efficient deployment of large AI models on physical systems.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by NVIDIA Blog.