Veo is Google's family of video generation models, which turn text or images into short video clips; newer versions also generate sound.
In plain words
Google introduced Veo in May 2024 as a model for high-definition 1080p video in many visual styles, aiming for people and objects that move realistically. Veo 3, announced at Google I/O in May 2025, generates audio too. Veo 3.1 makes 8-second clips at up to 4K with sound, in landscape or portrait. On 22 October 2026 Google closed the preview of Veo 3.1 that ran in its Gemini API, pointing developers to a newer model, Gemini Omni Flash, or to the generally available models on its Gemini Enterprise Agent Platform.
A quick example
Google's own docs show a prompt about two people studying a mysterious wall drawing by torchlight, with a line of speech. Veo 3.1 returns an 8-second clip that includes the scene, the spoken words and background sound.
Where you'll meet it
- The Gemini app, starting with Veo 3.
Related
Sources
- DocsVeo and Imagen 3, Google · read 29 Sept 2026
- DocsGenerate videos with Veo 3.1, Gemini API docs · read 2 Nov 2026
- DocsVideo generation in the Gemini API, Gemini API docs · read 5 Nov 2026
- DocsGoogle I/O 2025: all our announcements, Google · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 5 Nov 2026. Spotted an error? Report it.