we_are_coded.by CODE · The world, decoded
БГ

Google · Video

Gemini Omni 1.1 arrived free in Google Vids for anyone with a Google account

From 23 September anyone with a personal or Workspace account can generate HD video scenes in Google Vids with the Omni 1.1 Flash model, without paying. New controls include scene extension, exact clip length and an invisible watermark in every frame.

Source
Google Blog
Published
The event
Checked on
Responsible editor
Cvetelin Ivanov
How we work
MethodCorrections

A month after Google released Omni 1.1 to developers and paying subscribers, the model has landed in a free tool for everyone.

The road here is: first to those who pay and build, then to everyone else. And this time the free thing is HD video.

Fact

The facts: on 23 September 2026 Google announced that anyone with a Google or Google Workspace account can generate video scenes for free in Google Vids with the Gemini Omni 1.1 Flash model. Access is through vids.new on a desktop computer, by choosing Create AI videos. New controls cover scene extension that keeps the lighting, the characters' appearance and the environment consistent, setting an exact clip duration, and generating in full HD (1080p), as well as upscaling already generated clips to 1080p. Every clip made with Omni 1.1 in Google Vids contains the invisible SynthID digital watermark. Google says Gemini 3.8 Flash-Lite voiceover in more than 100 languages is coming to Vids soon. Paid Google AI and Workspace plans are offered for more generation.

What it does not say

How many clips the free tier includes. Google gives no number, it only says paid plans offer more. Nor does it say whether access is the same in every country. We will try it from Bulgaria before saying it works here.

Otherwise the direction is clear. A small business that used to pay someone for a 15-second promo video can now try it alone. The result will not look like a studio's; whether it is enough for a social media storefront, the test will show.

Free video will not take the director's job. It will take the template's job.

The control worth trying first is scene extension with your own footage, because it shows whether the model holds the character from shot to shot.

The visual is generated code-art, no third-party images.