Back to blog

Model updates

How to Make Videos With Gemini: Plans, Limits, Flow Credits and Steps

Gemini makes videos with Gemini Omni Flash, on a paid Google AI plan. Who can use it, how the limits work, Flow credit costs and the steps, read at Google's own pages on 2026-11-01.

Nov 1, 2026By Kevin Fan

Video examples here were generated on Cicadas and keep their model, settings and date. Cicadas earns money when you generate with these models. How we test

Test window:
2026-09-08 – 2026-10-12
Clips recorded with these models:
17
Models:
Gemini Omni Flash 1.1, Veo 3.1 Fast
Their provider cost:
≈$12.30
Selection:
every clip we keep is listed with its settings, cost and request ID; failed or discarded runs are listed in the test record.
How to Make Videos With Gemini: Plans, Limits, Flow Credits and Steps

In short: Gemini now makes videos with Gemini Omni Flash, which Google says replaces Veo in the Gemini app. You need a paid Google AI plan (Plus at $4.99 a month, Pro at $19.99 or Ultra from $99.99 in the US), you must be 18 or over, and you open Create video in the sidebar at gemini.google.com. Google gives no fixed number of videos a day: the app uses compute-based limits that refresh every 5 hours up to a weekly cap, and video uses the most of them. Google Flow, Google's filmmaking tool, is the place to count clips: it prices every generation in Flow credits, from 7 credits for a 4-second Omni clip at 720p, and gives 50 credits a day on top of a plan's monthly credits. Every Gemini video carries SynthID, Google's invisible watermark.

We make Cicadas, which runs Gemini Omni Flash 1.1 and Veo 3.1 Fast through fal.ai on the Gemini Omni Flash page and the Veo 3.1 page. Cicadas is not affiliated with Google. Every Google fact below was read on Google's own help, plan or developer pages on 2026-11-01; where Google does not state something, we say so.

1. Which model makes Gemini videos now

There are three Google routes, and they no longer run the same models.

  • The Gemini app (gemini.google.com and the mobile app): Gemini Omni Flash. Google's Gemini Omni page calls it "Gemini Omni 1.1 Flash" and says it replaces the Gemini app's previous Veo 3.1 model. It lists 10-second videos, native audio, photos to video from up to 5 images, scene extensions, video-to-video editing, multi-turn editing and an avatar of yourself.
  • Google Flow: both families. Flow's models page lists Veo 3.1 Lite, Fast and Quality next to Gemini Omni Flash 1.1, which Flow runs at 720p or a 360p draft quality.
  • The Gemini API, for developers: Gemini Omni Flash. Google's video generation docs, updated 2026-10-22, say the Veo 3.1 preview models in the Gemini API were shut down on October 22, 2026 and point developers to gemini-omni-1.1-flash.

Some of Google's plan FAQ text still mentions Veo 3.1 for Pro and Ultra. The video help page and the Gemini Omni page are the newer statements, and both describe Omni.

If you want to see how the two models differ on the same prompt, our Omni 1.1 vs Veo 3.1 page has six recorded clips side by side.

2. Who can make videos in Gemini

Google's video help page sets four conditions for a personal account:

  1. A Google AI plan. The feature table on Google's limits page gives video generation to AI Plus, AI Pro and AI Ultra and not to accounts without a plan. On the US plan page, Plus is $4.99 a month, Pro $19.99, and Ultra $99.99 (5x the Pro limits) or $199.99 (20x); these are the prices on the US page. The US plan page also advertises a free year of Pro for college students; we did not check who qualifies.
  2. Age 18 or over. The help page says the feature is not available to users under 18.
  3. Signed in. You must be signed in to Gemini. Work and school accounts need a qualifying Workspace license instead of a personal plan.
  4. A supported country. Google says AI Plus is available in more than 160 countries and territories and AI Pro and Ultra in more than 150. The lists we read include the United States, India, Japan, South Korea and Türkiye. They do not include mainland China, where Google's Gemini web app list says Workspace accounts only.

Uploading a video to edit it is not available in the EEA, Switzerland, the United Kingdom and some US states, according to the same help page; making a video from text or photos is not limited that way.

3. How many videos you can make: the limits, per plan

The Gemini app does not count videos. Google's limits page, read on 2026-11-01, describes compute-based limits that factor in the prompt, the feature and the length of the chat; the limit refreshes every 5 hours until you reach a weekly limit. Relative to an account without a plan:

  • AI Plus: 2x the standard limits.
  • AI Pro: 4x the standard limits.
  • AI Ultra: 5x or 20x the AI Pro limits, depending on the subscription.

The same page rates video generation as a "high" cost request and says video generally uses the most of your allowance. Google gives no number of videos per plan, so any "3 videos a day" figure you read elsewhere is not from Google. To see where you stand, open gemini.google.com and choose Settings, then Usage limits. Gemini warns you when you are close to the limit and tells you when it refreshes, and Google's plan page says paid subscribers can extend the limit by buying AI credits.

4. How to make a video in the Gemini app, step by step

These steps follow Google's help page for computers; the same help page has separate tabs for Android and iPhone.

  1. Go to gemini.google.com and sign in with the account that has the Google AI plan.
  2. Open the sidebar at the top left and choose Create video.
  3. Optionally pick a template.
  4. Type what should happen. Name the subject, the action, the camera and the light, and say what you want to hear: Gemini can make sound with the video.
  5. To start from your own media, choose Add image or Add video. Google allows one video and up to 5 images per request, and you must have the right to use them.
  6. Choose the aspect ratio before you generate. Text prompts default to landscape; with an uploaded image or video, the video takes the shape of what you uploaded.
  7. Submit. Google says it can take a few minutes, and you cannot use that chat while it works, so start a new chat if you want to keep talking.
  8. To change the result, describe the edit in the same chat: remove or replace an object or character, change the camera angle or the scene.
  9. To keep it, click Share under the video, then Download video or Share on YouTube.

For prompt structure, the brief and settings in our Gemini Omni 1.1 Flash guide can help in the app too.

5. Google Flow: credits per clip

Flow is the route to use when you want to count clips and pick the model. Its credits page, read on 2026-11-01, says every user gets 50 Flow credits a day, and that the daily refresh starts from your first generation and does not roll over. Plans add monthly credits:

  • AI Plus: 200 extra Flow credits a month.
  • AI Pro: 1,000 a month.
  • AI Ultra: 10,000 a month on the $100 plan, 25,000 on the $200 plan.

Credit costs per generation, as listed:

  • Gemini Omni Flash at 720p: 7 credits for 4 seconds, 10 for 6, 12 for 8, 15 for 10.
  • Gemini Omni Flash at 360p: 4, 5, 6 or 7 credits for the same lengths.
  • Editing a video with Gemini Omni Flash: 40 credits.
  • Veo 3.1 Lite: 10 credits for non-Ultra subscribers, 5 for Ultra.
  • Veo 3.1 Fast: 20 credits for non-Ultra subscribers, 10 for Ultra.
  • Veo 3.1 Quality: 100 credits, 8-second videos.

So the 50 daily credits cover seven 4-second Omni clips at 720p, or two Veo 3.1 Fast clips, if each request makes one video; Google notes that a single Flow request may generate 2 videos and charges per generation. On the Pro plan, $19.99 for 1,000 monthly credits, an 8-second Omni clip at 720p works out at about $0.24 by our arithmetic.

Two catches from Flow's own help. Its getting-started page lists a Plus, Pro or Ultra plan and age 18 among the requirements, although the credits page describes 50 daily credits "with no subscription". And without a plan, Flow says you may not be able to generate videos from roughly 2 to 5 PM UTC. Flow's region list, like the plan list, includes the United States, India, Japan, South Korea and Türkiye and not mainland China.

If you want Veo 3.1 without a Google plan, our Google Flow alternative compares Flow's prices with paying per clip.

6. Watermark and SynthID

Google's Gemini Omni page says every video generated in the Gemini app is embedded with SynthID, Google's invisible watermark for AI-generated content, and that you can upload a file to Gemini and ask whether it was made with Google AI. In Flow, all Veo, Omni and Nano Banana outputs carry SynthID. Flow also has a "Visible watermarking" switch in the menu under your profile picture, and Google applies a visible watermark automatically for users in India, South Korea and Vietnam who are not on the Google One AI Ultra 20x plan.

7. Gemini app, Flow, API or another site: which fits you

  • You already pay for Google AI and want a quick clip from a sentence or a photo: the Gemini app. Ten seconds, sound, edits by chat.
  • You want to choose the model, the length and the price of each clip: Google Flow, where every generation shows its credit cost before you run it.
  • You are a developer: the Gemini API. Its pricing page lists gemini-omni-1.1-flash on the paid tier only, at $17.50 per million video output tokens, which Google equates to about $0.10 a second of 720p video.
  • You do not want a Google plan, or you want Omni next to other models on one account: Cicadas. Gemini Omni Flash 1.1 is 100 credits for 4 seconds at 720p (about $0.99 on the $9.90 first pack of 1,000 credits), 200 for 8 seconds (about $1.98) and 300 for 8 seconds at 1080p (about $2.97), always with sound. Veo 3.1 Fast is 100 credits for 4 silent seconds and 300 for 8 seconds with sound. Per clip that costs more than a Google plan you would use anyway; there is no free tier, the quote shows before you generate, and a failed generation returns its credits. Clips are 4, 6 or 8 seconds here, not 10. Packs and plans are on the pricing page.

The cover of this post is omni-neon-umbrella, a 4-second vertical 720p clip with sound that we recorded with Gemini Omni Flash 1.1 on the Gemini Omni Flash page: a figure under a clear umbrella walks away down a rainy neon street. The shop signs came back with made-up lettering although the prompt asked for no text.

How we checked

On 2026-11-01 we read Google's Gemini Apps help pages "Generate videos with Gemini Apps", "Gemini Apps limits & feature availability" and "Where you can use the Gemini web app"; the Gemini Omni page and the US Google AI plans page; Google Flow's help pages "Manage your Google Flow credits", "Learn about Google Flow models & supported features", "Get started with Google Flow" and its supported-regions list; and the Gemini API video generation docs and pricing page. We did not generate in the Gemini app or Flow for this post, and we quote no Gemini video count per day because Google publishes none. Cicadas prices come from the per-second rates in our generator at the $9.90 first pack. Method: how we test.

Sources: Generate videos with Gemini Apps, Gemini Apps limits & feature availability, Where you can use the Gemini web app, Manage your Google Flow credits, Google Flow models & supported features, Get started with Google Flow, Google Flow supported regions.