Evergreen Tools
|
← Back to AI Tools

Gemini Omni Flash Video World Model

A video world model announced at Google I/O on 19 May 2026, whose first release Gemini Omni Flash generates video from any mix of text, image, audio and video input and supports multi-turn conversational editing with character and scene consistency, live in the Gemini app, Flow and YouTube Shorts

Tool Interface

Interactive tool will be available soon

Features

  • ✓ Create video from any input: freely combine text, images, audio and video as source material
  • ✓ Conversational editing: ask for a night scene or a different jacket and refine the same shot turn by turn
  • ✓ Grounded in Gemini's understanding of physics, history, science and culture for more plausible scenes
  • ✓ Generates native audio alongside the picture and embeds a SynthID watermark
  • ✓ Available across the Gemini app, Google Flow, YouTube Shorts and the Create app

How to Use

  1. Sign in with an eligible Google account in the Gemini app or Google Flow
  2. Upload images, video or audio, or type a description to build the source material
  3. After the first generation, refine it in plain language turn by turn — lighting, props, camera moves
  4. Export the finished clip, or use and remix it directly inside YouTube Shorts

FAQ

What is Gemini Omni Flash?

Gemini Omni Flash is the first model in the Gemini Omni family, announced by Google DeepMind at Google I/O on 19 May 2026. Google positions it as a step toward a world model that can create and edit anything from any input, starting with video.

How does it differ from video models like Veo?

Veo focuses on generating high-quality video clips, while Omni ties generation, understanding and editing together: it reads text, images, audio and video at once, edits the same scene conversationally across turns while holding consistency, and combines Gemini's intelligence with Google's generative media models.

Can I use it for free?

Google AI Plus, Pro and Ultra subscribers can use it in the Gemini app and Google Flow, and YouTube Shorts and YouTube Create users got free access from launch week, making it one of Google's broadest AI video rollouts.

Will generated video be watermarked?

Yes. Every video generated with Omni is embedded with a SynthID watermark so it can be identified as AI-generated, part of Google's push for AI content transparency, a standard several third parties have also adopted.

What are the known limitations?

Per Google DeepMind's model card, Gemini Omni Flash still struggles with complete consistency across many edits, complex motion scenes and rendering perfectly accurate text. The card is updated as the model improves.