Back to News

Google DeepMind unveils agentic video understanding with Gemini

#gemini#video understanding#agents#google deepmind

Google DeepMind announced agentic video understanding capabilities powered by Gemini, enabling AI agents to interpret and act on video content in real time. The system integrates multimodal reasoning with interactive video analysis, allowing agents to answer questions, track objects, and execute tasks based on visual streams. This advancement targets applications in robotics, augmented reality, and automated video surveillance.

Coverage timeline

  1. Google DeepMind Blog
    Introducing agentic video understanding with Gemini

Related stories

  • Google DeepMind unveils Gemini 3.8 Live with Live Avatar for Enterprise

    Google DeepMind announced Gemini 3.8 Live, a real-time conversational AI update featuring Live Avatar, an animated persona that lip-syncs and displays facial expressions during interactions. The feature is currently available only to Gemini Enterprise customers and supports 97 languages without degrading video fidelity or introducing visual drift.

    3 reports · 3 sources

  • YouTube launches AI-powered custom feeds via Gemini

    YouTube has introduced custom feeds that let users describe the videos they want in their own words, with Gemini generating a personalized feed based on the request. The feature aims to give users more control over their content discovery.

    1 reports · 1 sources

  • Google's Gemini app launches on Windows

    Google has made its Gemini app available for Windows, expanding access to the AI assistant beyond mobile and web platforms. The app is now downloadable on Windows devices, bringing Gemini's features to a broader desktop audience.

    1 reports · 1 sources56 points · 45 comments

  • Google DeepMind launches Gemini Omni 1.1 Flash video model with scene extension, 4K upscaling

    Google DeepMind announced Gemini Omni 1.1 Flash, a video production model with scene extension, first/last frame interpolation, and 4K upscaling. It reads up to 10 seconds of prior video context, extends scenes in 10-second increments up to 40 seconds total, and offers a 360p draft mode with up to 60% higher throughput and about one-third the cost of 720p.

    4 reports · 4 sources296 points · 233 comments

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe