Google just leveled up its AI game big time. Gemini Omni blends the model's sharp reasoning with serious creative power, turning any mix of images, audio, video, and text into high-quality videos.
Last year Nano Banana already let millions restore old photos, sketch ideas, and visualize wild concepts. Now Omni takes it further by starting with video creation and editing.
Edit Videos Like Magic
Forget complicated software. Talk to Omni in plain English and watch your clips transform. Instructions build on each other, characters stay consistent, physics feels real, and the story remembers what happened before.
Want a sculpture made of bubbles? Or turn a mirror touch into liquid ripples with reflective arms? Just say it. You can reimagine actions, add objects, or change entire scenes across multiple turns.
Smarter Than It Looks
Omni doesn't just copy styles. It reasons with Gemini's real-world knowledge of physics, history, and culture. Rolling marbles with perfect gravity? Claymation protein folding? Alphabet items with custom lower thirds and music? Done.
You can feed it references, images, existing clips, or audio, and get one smooth output. Early avatar support lets you create videos with your own voice and look.
All videos carry an invisible SynthID watermark for easy verification in the Gemini app, Chrome, or Search.
