Google Gemini Video 2026: Omni, Veo, Limits and Pricing

Last Updated: November 11, 2026

Summary: Google Gemini makes videos with Gemini Omni, the model that replaced Veo 3.1 in the Gemini app, and it can also watch and analyze videos you upload or find on YouTube. Video generation needs a Google AI Plus, Pro or Ultra plan (or a qualifying Workspace license), clips run 3 to 10 seconds (up to 15 seconds on AI Ultra), and every clip carries an invisible SynthID watermark. Developers pay $0.07 per second of 720p video for Gemini Omni 1.2 Flash through the Gemini API.

Much "Gemini video" advice online still describes Veo 3 and 8-second clips. That changed in 2026: Google launched Gemini Omni in May, moved the Gemini app onto it, and released Gemini Omni 1.2 Flash on November 5 with 15-second clips and lower pricing.

This guide covers making clips (in the Gemini app, Flow, Google Vids and the API) and using Gemini to understand video you already have. Every limit and price was checked against Google's own pages on November 11, 2026.

Google Gemini video means two things: generating new clips from prompts, images or footage, and asking Gemini about existing videos. Both run in the Gemini app, with separate limits and, for developers, separate pricing.

Most searches are about the first job, so this guide starts there. New to the app? Our guide on how to use Gemini covers the basics.

No. The Gemini app now generates video with Gemini Omni, and Google's Gemini video generation page describes Veo 3.1 as "the previous" model. The Gemini Apps Help Center calls Gemini Omni 1.2 Flash Google's "latest AI video generation model."

Veo has not disappeared entirely. Veo 3.1 models are still listed on Google Cloud's pricing page, and Google Photos in the US still lists Veo 3 for photo-to-video. But in the Gemini app and Flow, Omni does the work. Here is how the switch happened:

Google's Gemini Omni launch post describes a model that takes text, images, audio and video as inputs and keeps characters, physics and scenes consistent across conversational edits.

To make a video in Gemini, sign in on an eligible plan, describe the scene, and Gemini returns a clip you can edit, download or share to YouTube. Google's Generate videos with Gemini Apps help page lays out the options:

Gemini also supports avatars: type @ and your Google username to put your own avatar into a clip.

Gemini video generation is included in Google AI Plus, AI Pro and AI Ultra; the free Gemini tier does not include it. Google's Google AI plans page lists Gemini Omni Flash as the video model on all three paid plans, with different access levels and monthly Flow credits:

Google does not publish a fixed daily video count for personal Google AI plans. The Gemini Apps limits page says usage is compute-based, that "video generally requires the most usage," and that limits refresh every five hours until you reach a weekly cap.

Workspace accounts get firm numbers. Google's Gemini for work accounts page lists "up to 3 videos / day using Omni" on Business Standard, Business Plus, Enterprise Standard, Enterprise Plus and the AI Expanded Access add-on, and up to 5 videos a day with AI Ultra Access. Google warns limits "may change without notice."

Gemini app videos run 3 to 10 seconds, or up to 15 seconds with AI Ultra, and include generated audio on request. Google does not publish the output resolution for clips made in the Gemini app, so treat any specific number you see elsewhere with caution.

The developer versions are better documented:

Omni 1.2 specs come from Google's Gemini Omni 1.2 Flash announcement, which admits the model still struggles with fast sports action and complex physics. Veo specs come from Google's Veo documentation.

Google Flow is Google's separate AI video creation tool, running the same Gemini Omni Flash model as the Gemini app. Google positions Flow for directing scenes and multi-turn edits; the Gemini app suits a quick single clip. Flow draws on your plan's monthly credits, but Google's plans page does not say how many credits one clip costs, so check inside Flow before budgeting.

Google Vids now lets anyone with a personal Google account or Workspace account generate Omni clips at no cost, with 1080p scenes and templates, according to the Google Workspace blog; Omni 1.2 adds 15-second clips there at no additional cost. YouTube Shorts and the YouTube Create app also offer Omni free. For teams that mainly need short social clips, Vids may be the cheapest place to start.

Gemini Omni 1.2 Flash costs $0.07 per second of 720p video, according to Google's November 5 announcement, which puts a 15-second clip at $1.05. The model is generally available in the Gemini API, AI Studio and Google Cloud.

Google Cloud bills Omni by token. Its generative AI pricing page lists video output at $17.50 per million tokens for Gemini Omni Flash and Omni 1.1 Flash, with 720p video counted at 5,792 tokens per second. That is about $0.10 per second for Omni 1.1, consistent with Google's "30% lower" claim for 1.2. Veo 3.1 is on the same page:

Google's Gemini API pricing page lists the Veo 3.1 figures per second of video, and there is no free tier for video generation. The Veo 3.1 Preview models in the Gemini API shut down on October 22, 2026, and Google's Veo docs point developers to Gemini Omni Flash or the generally available Veo models on Google Cloud.

For a full breakdown of Google's text and image model rates, see our Gemini API pricing guide.

Yes. Google's Upload and analyze files page says the Gemini app accepts up to 5 minutes of total video (1 hour on AI Pro and Ultra), with each file up to 2 GB.

Gemini also works with YouTube. Per Google's YouTube help page for Gemini, you can ask it to find public videos and summarize one. Keep Activity must be on.

Developers get more room. Google's video understanding docs say models with a 1 million token context window can process up to about 1 hour of video at high media resolution or about 3 hours at low resolution. Gemini samples 1 frame per second by default, uses roughly 100 tokens per second at low resolution, and accepts public YouTube URLs.

That makes analysis cheap. A 30-minute recorded sales call at low resolution is about 1,800 × 100 = 180,000 input tokens. At the $0.50 per million input rate Google lists for Gemini 3 Flash Preview, that is about $0.09 before output tokens.

The strongest business uses for Gemini video are short marketing assets, fast creative testing and turning footage into searchable text. Google's Vids announcement names product launches, local shops turning stock photos into promos, and marketing clips for landing pages, social posts and event signage.

Sameer's take: I've tried a few AI video tools. The biggest business mistake with AI is trusting output without checking it.

Every video made with Gemini Omni carries SynthID, an imperceptible watermark that identifies Google AI content. Google says anyone can check whether a video was made with Omni through the Gemini app, Gemini in Chrome and Google Search.

With Sora gone, Kling is the closer like-for-like rival to Gemini Omni. OpenAI's help center says the Sora web and app experiences were discontinued on April 26, 2026, and gave September 24, 2026 as the date the Sora API would be discontinued.

Kling figures come from Kling's own Kling 3.0 credit cost guide, dated July 28, 2026, which notes prices can change. Our Kling AI pricing breakdown covers its credits in detail, and our Sora pricing page explains what happened to OpenAI's plans.

Which should you pick? If your team already pays for Workspace or a Google AI plan, start with Gemini: video sits next to Vids and YouTube, and Workspace data protections apply. Test Kling if you need native 4K or a cheap standalone video subscription.

Can Google Gemini make videos for free?
Not in the Gemini app, which requires a Google AI Plus, Pro or Ultra plan or a qualifying Workspace license. Google Vids and YouTube Shorts offer Gemini Omni clips at no cost.

Does Gemini use Veo 3 to make videos?
No. The Gemini app switched from Veo 3.1 to Gemini Omni, and Gemini Omni 1.2 Flash is now Google's latest video model. Veo 3.1 is still sold on Google Cloud, but the Veo 3.1 Preview models in the Gemini API shut down on October 22, 2026.

How long can a Gemini video be?
Gemini app clips run 3 to 10 seconds, or up to 15 seconds with Google AI Ultra, and editing can extend a clip by up to 10 more seconds.

How many videos can I make with Gemini per day?
Google publishes no fixed daily count for personal plans; limits are compute-based, refresh every five hours and stop at a weekly cap. Workspace business and enterprise editions allow up to 3 Omni videos a day, or 5 with AI Ultra Access.

Are Gemini videos watermarked?
Yes. Every Gemini Omni video includes an imperceptible SynthID watermark, and anyone can check for it in the Gemini app, Gemini in Chrome or Google Search.

Google Gemini video in 2026 runs on Gemini Omni, not Veo. For most businesses, the practical path is simple: use Google Vids or the Gemini app on an existing Google AI or Workspace plan for short marketing clips, and move to the Gemini API at $0.07 per second when you need volume or automation.

Check limits and prices on Google's own pages before budgeting, because Google changed models twice this year, and review every clip before it goes public.

⭐ Get more AI Business Weekly in Google: add us as a preferred source on Google.

Written by Sameer Khan with AI assistance, fact-checked against the sources linked above.