#United States #San Jose #Linker Vision #Nvidia GTC #Video Reasoning AI

Linker Vision Unveils Revolutionary Video Reasoning AI at NVIDIA GTC 2026

In a groundbreaking move within the AI landscape, Linker Vision, an esteemed player in the realm of AI platforms, presented its cutting-edge Video Reasoning AI during the NVIDIA GTC 2026, held in San Jose, California. This innovative technology is tailored to enhance intelligent operations in smart cities and interconnected environments.

The Video Reasoning AI platform, which has already been successfully implemented in Taiwan and Vietnam, is designed to collaborate with Voxelmaps and Inchor to provide sophisticated digital twin traffic simulation capabilities. This exciting initiative aims to explore the potential of city operation agents that can perceive, simulate, reason, and take action within the urban framework of San Jose.

How Video Reasoning AI Works

Linker Vision’s platform operates by enabling contextual comprehension of live video streams, which is crucial for various applications, including traffic management, industrial surveillance, infrastructure oversight, and ensuring public safety. By utilizing NVIDIA's Metropolis Blueprint for Video Search and Summarization (VSS), the system is capable of analyzing large-scale video data. Furthermore, Linker Vision leverages NVIDIA's Cosmos™ open world foundation models for enhanced understanding and reasoning capabilities, converting visual data into structured, actionable intelligence.

Central to the functionality of this technology is a training pipeline that employs digital twins, in alignment with the NVIDIA Physical AI Data Factory Blueprint. This comprehensive architecture combines data curation, generation, and validation, enabling efficient scale. The Physical AI Data Factory Blueprint utilizes Cosmos WFMs, including Cosmos Transfer and Cosmos Evaluator, powered by Cosmos Reason, to generate extensive and diverse datasets from limited initial training data.

Scalable Solutions in Action

Linker Vision is actively engaging in trials of a Physical AI toolchain on the Azure platform, which integrates the aforementioned blueprint alongside several Azure services. This collaboration facilitates scalable model development and continuous refinement of operations in complex physical environments, yielding model-ready training datasets from raw inputs.

To effectively manage high-volume, real-time video streams, Linker Vision is in partnership with leading telecom operators dedicated to establishing AI grids. These grid infrastructures are geographically distributed and based on NVIDIA's reference designs. A significant partnership with AT&T utilizes edge sites that function as AI grid nodes, allowing video reasoning capabilities to be deployed closer to the cameras and sensors, thus ensuring lower latency and operational reliability.

Additionally, the collaboration with T-Mobile is focused on piloting distributed deployments backed by AI-RAN ready infrastructure powered by NVIDIA RTX 6000 PRO Blackwell Server Edition. One notable initiative in San Jose integrates 3D point cloud data, digital twins, and vision AI, aimed at optimizing real-time traffic flow and urban planning.

Linker Vision is not just a technology provider; it seeks to create a robust ecosystem by intertwining simulation, training, and deployment into a continuous AI lifecycle. Paul Shieh, the Founder and CEO, eloquently stated, “As digital twins operate everywhere, reasoning AI can scale with them—unlocking new opportunities for smart cities and critical infrastructure.”

This commitment by Linker Vision positions it at the forefront of advancements in AI, paving the way for smarter urban environments and more efficient operational processes across various sectors. As the world becomes increasingly interconnected, technologies like Video Reasoning AI will play a pivotal role in shaping the future of smart cities and urban management.