What is Google Veo?
Google Veo is Google DeepMind's video generation model for turning text and reference images into video with native sound. The official model page currently presents Veo 3.1 as the leading version. You can try it in the Gemini app and in Google Flow, with plan limits that vary. The Veo 3.1 Preview in the Gemini API was shut down on October 22, 2026.
What you can do with it
On DeepMind's Veo page, the current leading model is Veo 3.1, described as a video generation model designed for filmmakers and storytellers, with native audio. A typical path is: write a shot description, optionally add reference images of a character, object, or style, and generate a clip with sound. The same page lists controls for extending a clip from the last second of a shot, transitioning between a first and last frame, directing camera moves, defining an object's path, and adding or removing objects while preserving scale and shadows.
Google surfaces Veo in more than one product. The model page points to the Gemini app, Google Flow (an AI filmmaking studio), Google Vids, Google AI Studio, the Gemini API, and Gemini Enterprise Agent Platform. Flow's own site lists Veo 3.1 alongside Gemini Omni and Nano Banana, and it notes that features may vary by Google AI subscription tier, platform, and region, with an 18+ restriction. DeepMind also states a limitation: natural, consistent spoken audio, especially for shorter speech, is still an area of active development.
A first workflow to try
Start on DeepMind's Veo page and choose Try in Gemini or Try in Google Flow, using the Google AI plan you already have. In Flow or Gemini, write a short prompt for a single shot with a camera description and, if you want speech or sound, say so in the prompt. If your plan allows reference images, add one picture of a character or a style and generate a brief clip. Watch the result for prompt fit, physics, and whether spoken audio stays in sync. AppFor.ai has not generated a Veo clip in this profile; follow the on-screen steps and your plan's remaining Flow or video allowance.
Plans and access
Consumer access is bundled with Google AI subscriptions. The pricing card above repeats the US personal figures checked November 11, 2026. In short, Free does not list video generation, Plus and Pro add it (Pro's FAQ also mentions a limited Veo 3.1 Lite trial), and Ultra's FAQ lists video generation with Veo 3.1 and a much larger Flow credit grant. Google states Plus is available in more than 160 countries and territories, and Pro and Ultra in more than 150. Those lists are Google's; this profile does not treat any one country as confirmed or excluded beyond that.
For developers, the Gemini API video documentation, checked the same day, says the Veo 3.1 Preview (including veo-3.1-generate-preview, veo-3.1-fast-generate-preview, and veo-3.1-lite-generate-preview) was shut down on October 22, 2026, and tells developers to migrate to Gemini Omni Flash (gemini-omni-1.2-flash) or generally available models on Gemini Enterprise Agent Platform. If you are choosing Veo for an API workflow, read that doc before you plan around a Veo 3.1 preview ID.
What stands out
- Text-to-video and image-guided generation with native audio, including sound effects, ambient noise, and dialogue
- Reference ingredients for a scene, character, object, or visual style, including audio on ingredients
- Scene extension, first and last frame transitions, camera controls, motion paths, outpainting, and object add or remove
- Output in 1080p and 4K, according to the official Veo page
Before you choose
- Veo 3.1 as a named consumer feature is listed on Ultra, with a limited Veo 3.1 Lite trial on Pro; Plus and Pro also list generic video generation, so the exact model on each tier can differ.
- The Veo 3.1 Preview in the Gemini API was shut down on October 22, 2026.
- Feature availability in Flow varies by plan, platform, and region, and Flow is limited to users 18+.
- Spoken-audio consistency is a limitation DeepMind still flags. Prices above are US personal rates and can change.
Go to the source
Sources & further reading
Based on the official documentation below. Features and access can change; check the provider before making a decision.
Was this page useful?
Help us make our coverage more useful. This is feedback on the page, not a rating of the tool.
Go a little further