Create Your Own Story with Gemini Omni

Google DeepMind introduced Gemini Omni and its first model, Gemini Omni Flash, at Google I/O 2026. The multimodal model starts with video generation, combines text, image, video and audio references,

Google DeepMind unveiled Gemini Omni at Google I/O on May 19, 2026. The product is a new family of Gemini models that creates results from multiple types of input, initially focusing on video output. The first model announced alongside it, Gemini Omni Flash, was designed to handle video creation and editing by combining Gemini’s reasoning capabilities with generative media models.

Gemini Omni can take text, images, video, and audio together as references and weave them into a single video result. During production, it supports conversational edits that change a scene’s setting, camera angle, style, and details through natural language, while continuing to incorporate previous instructions. Google explains that it uses Gemini’s knowledge of physical phenomena and history, science, and culture to create videos with greater context.

At launch, Gemini Omni Flash was available to Google AI Plus, Pro, and Ultra subscribers through the Gemini app and Google Flow, and was also rolled out sequentially to YouTube Shorts and the YouTube Create app. Every video generated with Omni includes an invisible SynthID digital watermark, and its generation status can be verified in the Gemini app, Gemini in Chrome, and Google Search.

Maru