How to use Veo 3.1 for free: access, credits & steps

You can try Veo 3.1 Lite in Google Flow without a paid Google AI plan if your account and region are eligible. Google lists 50 daily credits and charges most accounts 10 credits per Lite clip at 720p. That covers up to five attempts if you spend the full balance on Lite and free video generation is available.

Watch Veo 3.1 examples

Google DeepMind’s Veo 3.1 demonstration introduces audio and creative controls. These are vendor-selected examples, not our side-by-side test.

Source: Google DeepMind: Veo 3.1 - Designed to empower creatives

This guide starts with that access check, then walks through selecting the model and making one clip in Flow. You’ll see how credits, image inputs and basic troubleshooting fit the trial. Begin with the controls and balance shown in your own account, then generate a representative short scene before planning a larger workflow around it.

Quick answer

Open Google Flow, check your daily credit balance and select Veo 3.1 Lite. A four-, six- or eight-second 720p generation costs most accounts 10 credits. On a 50-credit daily balance, that covers up to five attempts before other use.

Test a current video model

Run one representative prompt or source image and compare the accepted output before choosing a paid workflow.

Try AI Video Generator

Where can you use Veo 3.1 for free?

Select Veo 3.1 Lite in Flow and check your balance. Free access covers basic generation on eligible accounts, although video can be unavailable during peak demand.

Google's Flow credit guide lists 50 daily credits without a subscription. Google’s getting-started page explains the region and verified-age requirements and which features need a paid plan. Open Google Flow in the account you intend to use and check its model picker and balance.

Flow requires an eligible account, age verification for users 18 or older, and a supported region. If your account is blocked, follow the account or region guidance shown by Google rather than assuming that a new prompt will fix it.

Google AI Studio and the Gemini API are a separate route. Google's API pricing page is for developer billing. Google shut down the Gemini API’s Veo 3.1 Preview models on October 22, 2026. If you need an API, confirm an available model and endpoint before budgeting. Your Flow daily credits apply inside Flow.

How to generate your first Veo 3.1 clip in Flow

Open Flow and sign in. Check your profile for the available credit balance.
Create or open a project. Choose text-to-video for a written scene, or frames-to-video if you have an image to animate.
Open the generation settings and select Veo 3.1 explicitly. Flow also offers other models; the product name alone does not identify the model producing your clip.
Choose the available aspect ratio and duration, then check the displayed credit charge and number of outputs.
Enter one clear shot description. Generate, review the result, and download a usable take before spending credits on more versions.

Google documents the controls in its Flow creation guide. Menu labels and available combinations can change.

How many free Veo videos can you make?

Here is what a 50-credit balance can cover at Google’s published non-Ultra rates, before other spending. Lite is the ongoing Veo option in Flow. Google lists Fast and Quality for retirement on December 5, 2026, so their rows apply only while the models remain available.

Veo mode	Credits per generation	Generations that fit within 50 credits
Veo 3.1 Lite	10	5
Veo 3.1 Fast	20	2, with 10 credits remaining
Veo 3.1 Quality	100	0

Google bills each output. Unused daily credits do not roll over. Free video may be unavailable at peak times. See the credit rules.

Can you animate a photo or use reference images?

You can use a first frame, first and last frames, or eligible ingredients. Google’s model feature matrix lists eight-second Ingredients/References for Veo 3.1 Lite and Fast, while Quality lacks that workflow. Use Lite for a new Veo setup and check the active model after selecting an input.

Choose the input based on what must stay consistent:

A specific starting composition: use an image as the first frame.
A planned beginning and ending: use first and last frames where supported.
Recurring subjects or visual references: check Ingredients/References availability in the chosen mode.
An entirely new scene: start with a text prompt and evaluate the result before adding complexity.

For help preparing an input, see our reference-image workflow. If you want to animate an existing image in Magic Hour, open image-to-video and check the current model and credit estimate before generating.

A simple prompt to start with

Use this original example as a starting point, not a tested performance claim. For a fuller framework, see our guide to writing video prompts.

A ceramic coffee cup rests on a wooden café table beside a window. Steam rises gently. The camera makes a slow, small push toward the cup. Soft morning light, natural colors, one continuous shot. Quiet café ambience, no speech or music.

Try in Text-to-Video

Review the clip against the brief: Is the cup stable? Is the camera movement useful? Does the audio fit? If one part fails, revise that part while keeping the rest of the brief consistent. Saving the prompt and settings alongside each take makes later comparisons easier.

For a product ad, replace the cup with an approved product reference, keep the action modest, and add exact offer text in an editor. Do not rely on a generated label to reproduce packaging, prices, or regulated claims correctly.

Why isn't free Veo generation working?

Access or region message: check Google's eligibility requirements and your account type.
Insufficient credits: check the selected model and output count; a mode can cost more than the whole daily allowance.
Unsupported reference or editing option: consult the model feature matrix rather than assuming every Veo mode has identical controls.
Temporary capacity restriction: free availability can vary during peak hours. A subscription is a purchasing decision, not a required troubleshooting step.
Poor result: simplify the shot, review your source image, and decide whether another attempt is worth the displayed charge.

Downloads, watermarks and finishing the video

Access, credit rates and model retirement dates were checked against Google’s documentation on November 9, 2026. Check the settings in your account before you generate.

Google's credit documentation lists 1080p upscaling as unavailable to non-subscribers. Check the available download resolution before committing to a workflow that needs a particular delivery format.

Google says Flow outputs made with its supported generative models include invisible SynthID watermarking. Visible watermark behavior is separate and can depend on settings and region; review the current watermark explanation. “No visible logo” does not mean “no AI watermark.”

A generated take is only one asset in the finished video. Assemble the selected clips, review the audio, add accurate captions with a subtitle workflow, and watch the exported file. For broader tool selection, compare image-to-video generators or video generators with audio.

To compare Veo access with other current options, see the best AI video generators guide.

Frequently asked questions

No. Flow is Google's creation application; Veo is a family of video models available within it. Confirm the model instead of assuming every Flow output is Veo.

First identify what is blocking your project: access, output length, reference control, resolution or cost. Use our Google Flow guide for the broader product decision, then compare the same brief and delivery requirements across the tools you are considering.