Gemini Omni is Google's video model in the Gemini app: you describe a video, or give it photos or a clip, and it creates or edits the video while you refine it in conversation. To use it, open the Gemini app on a paid Google AI plan, ask for a video or start from a template, then keep chatting to change the scene, lighting, background or length. You need to be 18 or over.
Google describes Omni as Nano Banana for video, and that's a fair summary. Instead of learning an editing timeline, you say what you want changed and Gemini changes it. Omni has also replaced Veo in the Gemini app, which is why some older guides no longer match what you see.
Here's how to get good results from it, what it can and can't do, and the safety details that matter if you post AI video for a business.
What Gemini Omni can do
| Feature | What it means in practice |
|---|---|
| Create from anything | Blend text, photos and video into a new clip |
| Photos to video | Animate up to five photo references into one video |
| Native audio | Sound is generated with the video, not added afterwards |
| Scene extensions | Ask Gemini what happens next and it lengthens the video while keeping the story |
| Video to video editing | Swap the background, change the wardrobe or transfer a style while keeping your details |
| Chat editing | Swap characters, adjust lighting, stabilise footage or change the background by asking |
| Templates | Curated styles to start from in a single tap |
| AI avatar | A digital version of you that looks and sounds like you, so you don't upload your photo every time |
The current model on Google's page is Gemini Omni 1.1 Flash, which Google says replaces the previous Veo 3.1 model in the Gemini app. Each generated video is ten seconds long, so plan longer pieces as several clips or use scene extensions.
How to make your first video
- Open the Gemini app or gemini.google.com and sign in to an account with a Google AI subscription.
- Ask for a video in plain language, or start from a template to see what's possible.
- To animate photos, upload up to five and describe how they should move.
- Watch the result, then reply with one change at a time: lighting, background, pace or wardrobe.
- When the story needs more, ask what happens next to extend the scene.
- Download or share the clip once you're happy with it.
Prompts that work
The trick is to describe a shot, not a topic. Tell Omni what the camera sees, how it moves and what it sounds like.
A product scene
A close-up of hands arranging fresh flowers on a wooden counter in a small florist's shop.
Soft morning light, the camera slowly moves back to show the shop.
Gentle background music and the sound of the door bell.
Photos to video
Use these three photos of my bakery.
Turn them into one video that moves from the shop front to the counter to a tray of fresh bread.
Warm, natural colours and no text on screen.
An edit in conversation
Keep everything the same, but change the background to a sunny street outside and make the lighting brighter.
Using an AI avatar
The avatar is the feature most business owners ask me about. Google's FAQ says it's completely optional, and only you can use your avatar to create videos. You can keep uploading your own photos instead, or skip that step and use the avatar.
My advice: if you use an avatar for marketing, tell your audience. People trust a brand less when they find out later that the person on screen wasn't really talking.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Who can use Gemini Omni
| Question | Answer from Google |
|---|---|
| Do I need a paid plan? | Yes, a Google AI subscription is required |
| Age limit | 18 and over |
| Is it available everywhere? | Google says features vary by tier and geography |
| Is there a student route? | Google is offering college students a free Pro plan for one year that includes Omni access |
Omni also shows up outside the Gemini app. In Google Vids, the Edit videos start option transforms existing clips with Omni, which is handy if your team already makes videos there.
AI labels and SynthID
Every video made in the Gemini app carries SynthID, Google's invisible watermark for identifying Google AI content. You can also upload a file to Gemini and ask whether it was made with Google AI, and it'll check for SynthID. If a client asks whether a clip is AI-generated, that's a quick way to show them.
Where Omni falls short
- Ten-second clips mean longer videos need planning and stitching together
- Five photos is the ceiling for a single photo-to-video request
- Feature availability depends on your plan and country, so a guide may show options you don't have
- You still need to check every frame for odd hands, text and logos before you post
Tools that work with it
Omni sits inside the Gemini app alongside Nano Banana for images, and it powers editing in Google Vids. If you're comparing AI video tools, Runway, Kling, Sora and Pika are the names most creators test next to it.
Google's other video tool works from templates, and our guide to Google Vids walks through a first project.
My take
Omni is at its best for short social clips, product moments and turning a few good photos into something that moves. Treat it as a fast first draft and keep each request to one change. If you're on a tight budget, my guide to where to create AI videos for free covers the free routes, and the best AI video generators for YouTube compares the main options. And if you want help building AI video into your marketing, that's what I do in my one-to-one AI sessions.
Frequently asked questions
What is Gemini Omni?
Gemini Omni is Google's video model in the Gemini app. It creates and edits videos from text, photos and clips, and you refine the result by chatting.
Did Gemini Omni replace Veo?
Yes. Google says Gemini Omni replaces Veo in the Gemini app, and Gemini Omni 1.1 Flash replaces the previous Veo 3.1 model.
How long are Gemini Omni videos?
Google lists ten-second videos, and you can lengthen a story with scene extensions by asking Gemini what happens next.
Is Gemini Omni free?
No. Google says a Google AI subscription is required, you must be 18 or over, and features vary by tier and geography.
Can someone else use my Gemini AI avatar?
No. Google says the avatar is optional and only you can use your avatar to create videos.
