NVIDIA has announced that NVIDIA XR AI is now available in public beta, giving developers a library for building spatially aware multimodal AI agents for augmented reality (AR) glasses and extended reality (XR) devices. By connecting XR device inputs with AI models, enterprise data, tools and accelerated computing, NVIDIA XR AI enables agents that can “perceive, reason and act in the flow of work.” The platform helps simplify the integration of multimodal perception, enterprise retrieval, reasoning models and agent orchestration, and provides a foundation for developers to build or connect skills and tools for enterprise XR applications. NVIDIA stated that the platform brings together four core capabilities: real-world data ingestion from AR and XR devices; specialized tools and services including NVIDIA Metropolis and VSS for visual AI and video understanding, and NVIDIA NeMo Retriever for enterprise knowledge retrieval; AI model support including NVIDIA Nemotron reasoning models, NVIDIA Cosmos Reason and other compatible foundation models; and agent orchestration and runtime services for moving applications from prototype to production. The platform supports tool use, reasoning workflows and deployment across cloud, data center and edge environments. NVIDIA highlighted several companies exploring applications, including Siemens, which is exploring how NVIDIA XR AI and NVIDIA DGX Spark can help factory engineers find maintenance information, and Rana, an AutoBio company, which is introducing its LabOS system for hands-free scientific guidance.