The new version of Veo allows users to generate video and audio simultaneously, with full editing capabilities and adherence to the prompt
Google introduced Veo 3.1, the new update of its video AI that allows users to turn images into clips with prompt adherence.
The tool aims to compete with Sora 2 and offer content creators more creative control.
While Sora 2 targets realism, Veo 3.1 focuses on being a useful tool for creators seeking to edit and produce flexibly. According to Google, it keeps prompt consistency and allows users to work with multiple frames to generate more complex videos.
In the style of Nano Banana, Veo 3.1 ensures subject uniformity and facilitates editing with multiple images. The artificial intelligence generates the intermediate clip between the first and last frames, adding audio at the same time.
The new version includes the "Frames to Video" feature, which allows users to upload key frames and have the AI generate the rest of the clip. This provides the possibility to insert objects, subjects, or additional elements at any point in the timeline, enhancing user creativity.
Google also highlights that Veo 3.1 makes it easier to extend clips and add synchronized sound without the need for external tools. This way, the editing experience becomes more comprehensive and efficient for video creators.
Veo 3.1 focuses on professional and creative editing. The tool represents another branch within the Gemini ecosystem, alongside Gemini 2.5 Flash Image (Nano Banana), bringing editing options closer to professionals and enthusiasts.
