Google Updates Veo 3.1: AI Now Creates Vertical 4K Video from Images
Google has introduced an update for its video generation model, Veo 3.1. The main changes involve working with source images, support for mobile formats, and increased resolution for final clips.
A key improvement is the "Ingredients" tool. The system now more accurately analyzes reference images uploaded by the user (up to three) and creates video with more pronounced dynamics based on them. The model also better maintains consistency: characters and objects look the same across different scenes, which is important for creating short, coherent narratives.
Native generation of video in a vertical 9:16 format has been implemented for the first time. This feature is geared towards creating content for platforms like YouTube Shorts and TikTok without the need for subsequent cropping. The option is available both when working with text prompts and when using "Ingredients." An upscaling capability has also been added. The base resolution for generated video remains 720p, but users can increase it to 1080p or 4K for a sharper picture.
The updated features are already integrated into the Google Flow service and the Gemini app. They will soon appear in YouTube Shorts and the YouTube Create mobile editor. All videos created by Veo 3.1 contain invisible SynthID watermarks to indicate their artificial origin.
The update makes the tool more practical for content creators, focusing on improving control over the result, adapting to mobile platforms, and enhancing the technical quality of the final video.
-
Microsoft's Cloud Gaming Coming to New TCL Google TV Sets -
Google TV's AI Gains New Features: From Volume Adjustment to Video Creation -
Google Pushes Back Full Assistant-to-Gemini Switch Until 2026 -
Google Now Lets You Change Your Gmail Address Without Losing Your Account -
Google Photos and Gemini AI Coming to Samsung TVs in 2026
