Google announced an update to its AI tool Veo 3.1 Ingredients to Video, which now allows creating videos based on up to three reference images. This enables users to more accurately control the appearance of videos by selecting characters, backgrounds, and textures for the video. The new features provide better visual consistency between the source images and the final video, which is especially useful for those aiming to achieve a specific result with minimal prompts.
Veo 3.1 has gained support for vertical format, making it convenient for creating content for mobile platforms, particularly for YouTube Shorts. Additionally, the ability to enhance video resolution to 1080p and 4K has been added, allowing professionals to obtain clearer and higher-quality materials for editing or publishing.
The update is already available in the Gemini app, on YouTube, in Flow services, Google Vids, via Gemini API, and Vertex AI. Both categories of users — regular creators and business representatives — can apply the new capabilities to create content of varying complexity and quality.
Improvements include enhanced consistency of character and object appearance across different scenes and detailed customization of backgrounds and textures. Google has also implemented SynthID digital watermarks in the created videos, promoting transparency and content origin verification for users and platforms.
