Google will now let users create videos from images as it brings a new image-to-video generation feature within Gemini, building upon the global availability of its Veo 3 video model to subscribers last week. This new tool is powered by the company’s latest video generation model, Veo 3.
Users can now upload an image, describe how they want it to move or sound, and Gemini will create an 8-second MP4 video at 720p resolution. The tool is available starting July 10 on the Gemini website (gemini.google.com) for users subscribed to the Pro ($19.99/month) and Ultra ($249.99/month) plans. Meanwhile, a mobile version of the feature is expected to launch soon in select regions.
Each video generated through this feature includes a visible ‘Veo’ watermark, showing it was created using Google’s AI. It also includes an invisible watermark known as SynthID, which allows the video to be identified as AI-generated even if it is modified or shared elsewhere. According to the company, it has put strong safety measures in place to prevent harmful or misleading use of this tool. Along with watermarking, the tech giant claimed that it conducted extensive internal testing (known as ‘red teaming’) to identify and fix possible risks before release.
To create a video from a photo in Gemini, users need to select the ‘Videos’ option from the prompt box, upload a photo, and enter a description of the desired scene and audio. Gemini will then generate a short animated video from the still image. Once the video is ready, it can be downloaded or shared directly.
“From reimagining fairy tales through the eyes of a modern influencer, to ASMR videos exploring what it would sound like to cut through a piece of cooling lava, your imagination is the limit when you create videos with Gemini,” the company said in a blog post.
The technology behind this feature is built on Veo 3, developed by Google DeepMind. The tech giant stated that since its release in May 2025, Veo 3 has been used to create over 40 million videos across various Google products, including Flow (a standalone AI video tool).
However, despite its rapid advancements, Google Gemini has faced several controversies. Most recently, concerns were raised over the app accessing services like WhatsApp and Messages without clear user consent, sparking privacy debates. Earlier, the platform was criticized for releasing updates without full safety reviews, and it had already drawn backlash for politically biased responses and racially inaccurate image generations. Meanwhile, Google is not alone in exploring the image-to-video generation feature. Earlier, in 2024, the social media powerhouse, Meta, introduced its own tool called ‘Movie Gen’, which transforms static images into short video ads by automatically adding motion, transitions, and background audio.
