Photos and video
How to Use Gemini to Create Videos
Most articles on this are describing a product that no longer exists in the Gemini app. Veo was the video model; Google replaced it with Gemini Omni in May 2026, and Omni is a paid feature. Here is what it actually does, what it costs, and where Veo still lives if that is what you were looking for.
GPTPrompts.AI Editorial
Hands-on with Gemini; every method here is one we use · Last updated July 2026
Affiliate link, we may earn a commission at no extra cost to you.
01 · Reality check
What Gemini actually does well here
Good at
- Ten-second clips generated from a text prompt, with native audio
- Turning still photos into video, using up to five reference images
- Video-to-video editing, where you upload a clip and describe changes
- Multi-turn editing, refining a clip across several messages
- Aspect-ratio control, and sharing straight to YouTube
Not the right tool for
- Anything on the free tier, since it needs a Google AI plan
- Users under 18, who cannot access it at all
- Clips longer than 10 seconds
- Uploading video for editing in the EEA, Switzerland, the UK or some US states
- Output without SynthID watermarking, which is applied to everything
02 · The method
Step by step
- 1
Check you can actually use it
You need a Google AI plan (Plus, Pro or Ultra) and to be 18 or over. There is no free path to video generation in the Gemini app.
- 2
Decide text-to-video or photo-to-video
A text prompt defaults to landscape. Uploading images inherits their aspect ratio, which is the easier route to a specific format.
- 3
Write for ten seconds, not for a film
You get a 10-second clip. Describe one shot or one simple movement rather than a sequence, or you will get a rushed, incoherent result.
- 4
Use reference images for consistency
You can supply up to five reference images. This is the most reliable way to control how a subject looks rather than describing it repeatedly.
- 5
Refine across turns
Omni supports multi-turn editing, so describe the change you want rather than regenerating from scratch.
- 6
Know the watermark and the region rules
Everything carries SynthID watermarking. If you are in the EEA, Switzerland, the UK or certain US states, uploading a video for video-to-video editing is not available to you.
03 · Use this now
Copy-paste prompt structure for a 10-second clip
Create a 10-second video of [one clear subject doing one clear thing]. Shot: [close-up / wide / overhead], camera [static / slow push in / pan left]. Lighting and mood: [describe]. Setting: [describe briefly]. Audio: [what should be heard]. Aspect ratio: [landscape / portrait]. Keep it to a single continuous shot with no cuts, because ten seconds is not long enough for a sequence. Do not add text overlays or people I have not described.
04 · Avoid these
Common mistakes
- Following a guide that says Veo powers video in the Gemini app. Gemini Omni replaced it in May 2026.
- Expecting video generation on the free tier, which does not include it.
- Writing a multi-shot script for a 10-second clip, which produces an incoherent result.
- Assuming video upload works everywhere. It is unavailable in the EEA, Switzerland, the UK and some US states.
- Planning to pass output off as un-generated footage. SynthID watermarking is applied to everything.
05 · Questions
Frequently asked questions
Can Gemini create videos?
Yes, through Gemini Omni, but not on the free tier. Google requires a Google AI plan and that you are 18 or over. It generates 10-second clips with native audio, can turn up to five photos into video, and supports video-to-video editing.
What happened to Veo in Gemini?
Google replaced it. Its own FAQ states that Gemini Omni is the latest video model and will replace Veo in the Gemini app, rolled out on 19 May 2026. Veo 3.1 still exists elsewhere: in Flow, the Gemini API, Vertex AI and Google Vids. So Veo is not discontinued, it is just no longer what powers video inside the Gemini app.
Can I generate video with Gemini for free?
No. Google states that to use the feature with a personal account you must have a Google AI plan. Many other Gemini features are genuinely free, including image generation and editing, Deep Research, Gemini Live and Gems, but video generation is not one of them.
How long can Gemini videos be?
Ten seconds, per Google's documentation. Write for a single continuous shot rather than a sequence. Google does not publish any per-day or per-month video count, only the relative usage multipliers between plans.
Is Gemini Omni a real product?
Yes. It is both the consumer video model in the Gemini app, marketed as Gemini Omni or Gemini Omni Flash, and a documented developer model in the Gemini API, where it is still in preview. Google announced it in May 2026.