Short answer: Google’s current Gemini Apps Help describes video generation with Gemini Omni, not a Veo 3 model selector. On a computer, choose Create video; on mobile, choose Menu → Videos. Enter a prompt and submit it. If you want to generate videos from software, that is a separate Gemini API workflow.
Create a video in the Gemini app
On a computer
- Go to Gemini and sign in.
- Open the sidebar and select Create video.
- Optionally select a template.
- Enter a description of the video and click Submit.
Google’s help page says you can add up to five images and one video when creating a video on a computer. It also says generation may take a few minutes.
On a phone or tablet
- Open Gemini and sign in.
- Tap Menu, then Videos.
- Optionally select a template, enter your prompt, and submit it.
The current help page identifies Gemini Omni as the app’s video generator; it does not document a Veo 3 selector. Older Veo 3 tutorials show an earlier interface, so treat their button names and playback steps as historical—not as a guarantee of what you’ll see now. Google’s current help does not specify a separate results library or exact playback control; follow the on-screen result after generation finishes.
Example prompt: “A small red canoe glides across a misty alpine lake at sunrise. Slow aerial camera movement, soft golden light, calm water, natural cinematic style.” This is an illustrative prompt, not a tested recipe or guarantee of a particular result.
What if you don’t see the video option?
Google says you must be signed in to Gemini Apps. For a personal account, video generation requires a Google AI plan. For a work or school account, it requires a qualifying Workspace license. The feature is currently unavailable to users under 18. Check that you’re signed in to the account with the relevant plan or license, and consult Google’s current help for the applicable Workspace plan details.
The help page’s computer instructions allow up to five images and one video as references. That upload limit is for the computer workflow; don’t assume the same limit applies on mobile. Google also notes a separate location restriction: uploading a video for editing is unavailable in the EEA, Switzerland, the UK, and some US states. That restriction is about video-edit uploads, not a general statement that video generation is unavailable in those places.
Gemini app or developer API?
The Gemini app is the interactive route: choose its video feature, describe what you want, and submit the prompt. The Gemini API is a separate developer route for integrating video generation into software. Google announced Gemini Omni 1.2 Flash as generally available across the Gemini API on November 5, 2026. Google lists 720p API video at $0.07 per second; that is API pricing, not the cost of a Gemini app plan.
Google’s general Gemini API setup guide says you need an API key. It says moving to the paid tier requires setting up Cloud Billing. Those general setup instructions do not, by themselves, establish every model-specific billing or access requirement.
Using the API for video
For Gemini Omni 1.2 Flash, use Google’s version-specific Cookbook guide for setup, generation, and understanding the video output. The current search result for Google’s API video guide identifies Omni 1.1, while Google’s announcement identifies Omni 1.2 as the newer generally available release; don’t copy an older model identifier into a 1.2 integration without checking the 1.2 guide.
This article does not provide a copy-and-run API example: the retrieved materials do not establish a complete, version-matched 1.2 request and response sequence here. In particular, verify the 1.2 request fields, how that request signals completion, how to retrieve the finished video, and the model-specific billing setup in the current Cookbook and API documentation before deploying. The Gemini API workflow is not the same as the Gemini app controls.
Bottom line
For interactive video creation, use Create video on a computer or Menu → Videos on mobile. Google’s current Gemini Apps Help calls the app model Gemini Omni, not Veo 3. If you’re building software, use the separate Gemini API documentation for the current Omni version rather than relying on older Veo preview examples.
Sources
Generate videos with Gemini Apps — Gemini Apps Help
Gemini Apps limits and upgrades for Google AI subscribers
Gemini Omni 1.2 Flash: 15-second video and lower pricing
Generate and edit videos with Gemini Omni 1.2 Flash — Google Gemini Cookbook
Generate and edit videos with Gemini Omni Flash — Google AI for Developers