Google has launched a new feature for generating videos based on text prompts in its Gemini and Whisk AI tools. Starting today, Gemini Advanced users can create videos using the Veo 2 AI model for video generation.
According to the company, Veo 2 is capable of generating eight-second videos in MP4 format with 720p resolution and 16:9 widescreen aspect ratio.
The company promises that the model has a better understanding of real-world physics and human movement, smooth character animation, realistic scenes, and fine visual details in a variety of themes and styles.
To create a video, users must select the Veo 2 model from the Gemini menu and enter a text description of the scene. The more detailed the description, the more control you have over the final video. According to Google, the new feature opens up a wide range of creative possibilities – from fairy tales to fantastical landscapes.
Below are the videos generated by Veo 2 model, provided by Google:
Users can share the created videos on TikTok and YouTube Shorts from your mobile phone by pressing the “Share” button.
In addition, Google has updated the experimental Whisk tool available in Google Labs with a new feature for animating images in videos – Whisk Animate. Users can now create videos from images as well as text queries.
Both features are available to Google One AI Premium subscribers on the web and mobile devices. Video creation will be rolled out in phases and will be available in all languages supported by Gemini.
For security purposes, all videos generated by Veo 2 will be digitally watermarked with SynthID, indicating their artificial nature. Google has also implemented additional measures to prevent the creation of content that violates the company’s policies.
Read more: Google opens access to its ‘smartest’ AI model Gemini 2.5 Pro for all users.