Nano Banana video isn’t a thing Google sells: Nano Banana makes still images, and a separate model called Gemini Omni makes the video. The two work well together, though. You can create or edit a picture with Nano Banana, then animate it in the Gemini app. This guide explains who does what, how to turn a photo into a short clip, what it costs in plan terms, and how to avoid the look-alike sites that use the Nano Banana name for their own video tools.
Nano Banana video: what is real and what isn’t
Nano Banana is Google’s image model. It does not output video. For video, the Gemini app uses Gemini Omni, which Google announced at its I/O event in May 2026. Google’s page says Omni replaces the earlier Veo 3.1 model inside the app. It makes clips of up to 10 seconds, with sound, from text, photos or existing video.
Key takeaways
- Nano Banana creates and edits images. Gemini Omni creates and edits video.
- In the Gemini app you can turn up to five photos into a video.
- Clips are up to 10 seconds and include native audio.
- A Google AI subscription is required, and the feature is for users aged 18 and over.
- All videos made in the Gemini app carry a SynthID watermark.
- Websites that sell “Nano Banana video” tools are usually not from Google.
How Nano Banana and Gemini Omni fit together
Think of it as a two-step studio. Nano Banana is the photographer and retoucher. Omni is the camera crew that makes the picture move. Google has pitched Omni as the video version of the Nano Banana idea, and both are designed to keep a person’s likeness consistent.
| Nano Banana | Gemini Omni | |
|---|---|---|
| Output | Still images | Video clips with audio |
| Input | Text or uploaded photos | Text, up to five photos, or video |
| Length | Not applicable | Up to 10 seconds |
| Best for | Posters, edits, product shots, portraits | Short social clips, animated scenes |
| Access | Free tier available, higher limits on paid plans | Google AI subscription needed |
If you’re new to the image side, start with our Nano Banana prompt guide. It teaches the wording that makes a still image clean enough to animate.
How to make a video from a photo in Gemini
- Prepare the image. Create it with Nano Banana, or choose a sharp photo of your own. Check the faces, hands and any text, because mistakes tend to carry into the video.
- Open the video option. In the Gemini app, choose the video creation tool. The exact menu label can change, so look for the video or Omni entry.
- Add your photo. You can use up to five photos as references.
- Describe the motion. Say what moves and how, in one or two sentences.
- Generate and review. Watch the full clip, then ask for a change in the same chat, such as a slower camera move.
- Download and label it. Keep the SynthID watermark in mind when you share, and add a note that the clip is AI-made.
How to write a good video prompt
A still image prompt describes a moment. A video prompt describes a moment plus change. A simple formula helps: subject, action, camera, setting, mood.
- Subject: who or what is in the shot.
- Action: what happens during the ten seconds. Keep it to one main action.
- Camera: slow push-in, static shot, gentle pan, overhead.
- Setting and light: place, time of day, weather.
- Mood or sound: calm, tense, busy street noise, soft music.
Example: “The woman in the photo turns toward the window and smiles. Slow push-in camera, soft morning light, quiet room with faint birdsong.”
Ten seconds is short. Ask for a single idea and the result usually looks better than a crowded one.
Prompt ideas to try
| Goal | Prompt |
|---|---|
| Product teaser | The mug on the table steams gently as the camera slowly pushes in. Warm kitchen light. |
| Travel clip | Waves roll onto the beach in the photo while clouds drift slowly. Static camera, golden hour. |
| Portrait | The man in the photo looks up from his notebook and nods. Shallow depth of field, office background. |
| Illustration | The cartoon fox in the picture hops across the log. Flat colours, playful music. |
Plans, limits and what to expect
Google’s pages say a paid Google AI subscription is required for Omni, and that what you get depends on your tier and where you live. The pages don’t list a fixed number of videos. One user reported on Reddit that four videos used an entire usage window on a Pro plan. That’s a single anecdote rather than an official limit, but it suggests video burns allowance faster than images do. Plan your prompts before you press generate, and don’t regenerate for tiny tweaks.
College students can get a free Google AI Pro plan for a year, according to Google’s page, which is worth checking if you qualify. Our Nano Banana Pro guide covers plan tiers for images.
Watch out for fake “Nano Banana video” sites
Search for the phrase and you’ll find sites using the name for their own paid generators. Some of them use Google’s models through an API. Others use different models entirely. Few are run by Google. Before you pay anyone, check these points:
- Is the address one of Google’s own (gemini.google or gemini.google.com)?
- Does the site ask for a card before showing what you get?
- Does it say clearly which model it uses and who runs it?
- Does it make big promises such as unlimited free video?
If the answers worry you, stay in the Gemini app. This site is independent and has no link to Google or those sellers. Our disclaimer says more.
Common problems with photo-to-video
| Problem | Why it happens | What to try |
|---|---|---|
| The face drifts during the clip | Small or blurry faces give the model little to hold on to | Start from a sharp, front-facing photo and keep the action simple |
| Strange hands or objects | Complex movement is hard to keep consistent | Ask for a calmer action, such as a head turn or a slow camera move |
| Text in the image warps | Letters often change when everything moves | Add the words in an editor after the video is made |
| Clip feels too short | Ten seconds is the cap | Make several clips and join them, or use a scene extension if your plan offers it |
| Request refused | Safety rules or a limit | Remove real names and sensitive content, then check your usage allowance |
Treat the first result as a draft. Professional video work is mostly iteration, and AI video is no different. Save the versions you like so you can compare them side by side.
Responsible use
Video that looks real can mislead. Google marks Omni videos with SynthID, and you can upload a file to Gemini to ask whether Google’s AI made it. Don’t use the tool to imitate a real person without their permission, to fake news or to harass anyone. Some observers have raised concerns about AI-generated “slop” and about realistic videos of real people, and the guardrails can feel tight. That is the trade-off for tools this capable.
If you can’t use Omni
- Make a slideshow-style reel. Create several Nano Banana images and cut them together in a normal video editor.
- Use a different video tool. Other AI video services exist, and you can feed them an image from Nano Banana. Check their terms and data rules first.
- Stay with images. Many posts work better as a strong still picture. See the Gemini and Nano Banana section for ideas.
Frequently asked questions
Can Nano Banana make videos?
No. Nano Banana is Google’s image model. Video in the Gemini app comes from Gemini Omni, which Google says replaces Veo 3.1 there. A common workflow is to create or edit a picture with Nano Banana, then animate it with Omni.
How do I turn a photo into a video in Gemini?
Open the video creation option in the Gemini app, add up to five photos, describe the motion in a sentence or two, and generate. Review the clip, then ask for changes in the same chat. A paid Google AI plan is required.
How long can Gemini Omni videos be?
Google’s page says up to 10 seconds, with native audio. For longer pieces, make several clips and join them in a video editor. Scene extensions may help where your plan offers them, but availability varies by tier and country.
Is Nano Banana video free?
Not in the Gemini app. Google’s pages say a Google AI subscription is required for Gemini Omni, with features varying by tier and location. Google lists a free one-year Pro plan for college students, so check whether you qualify.
Are sites that sell Nano Banana video tools official?
Usually not. Many third-party sites use the Nano Banana name for their own generators, and some use different models. Check whether the address belongs to Google, which model is used and what the pricing is before you enter payment details.
Do Gemini videos have a watermark?
Yes. Google says all videos made in the Gemini app include SynthID, an invisible watermark that identifies Google AI content. You can upload an image, video or audio file to Gemini and ask whether Google’s AI made it.
The bottom line on Nano Banana video
Nano Banana makes the picture, and Gemini Omni makes it move, but only for subscribers and only for ten seconds at a time. Treat any other “Nano Banana video” claim with caution. We update our guides as Google changes things, and you can see how we check them on our how we test page. For more tips, go back to the Google Edits homepage.
Independent publication. Not affiliated with or endorsed by Google. Limits and prices change; verify in your own account.