Google New VideoFX: Transforming Ideas into Stunning Videos

Updated: May 17 2024 23:56
At Google I/O 2024, Google Labs announced the launch of VideoFX, an experimental tool powered by Veo, Google DeepMind's most advanced generative video model to date. VideoFX is designed to empower creatives by transforming their ideas into captivating video clips with just a simple text prompt. This innovative tool is set to revolutionize the way people bring their stories to life.



At the heart of VideoFX lies Veo, a cutting-edge video generation model that produces high-quality, 1080p resolution videos spanning over a minute in length. Veo's advanced understanding of natural language and visual semantics allows it to accurately interpret text prompts and combine them with relevant visual references, resulting in videos that closely follow the user's vision.

Veo's capabilities extend beyond simple video generation. It can apply editing commands to existing videos, such as adding kayaks to an aerial shot of a coastline, and even supports masked editing for precise modifications. Additionally, Veo can generate videos based on a reference image, enabling users to create content that aligns with a specific style.


Prompt: A lone cowboy rides his horse across an open plain at beautiful sunset, soft light, warm colors

Veo builds upon years of generative video model work including Generative Query Network (GQN), DVD-GAN, Imagen-Video, Phenaki, WALT, VideoPoet and Lumiere, and also our Transformer architecture and Gemini.

To help Veo understand and follow prompts more accurately, Google have also added more details to the captions of each video in its training data. And to further improve performance, the model uses high-quality, compressed representations of video (also known as latents) so it’s more efficient too. These steps improve overall quality and reduce the time it takes to generate videos.



VideoFX introduces Storyboard mode, a powerful feature that allows users to iterate scene by scene and add music to their final video. This mode provides a perfect balance between quick explorations and creative control, making it an invaluable tool for filmmakers, educators, and aspiring creators alike.



Alongside the launch of VideoFX, Google Labs is also releasing updates for ImageFX and MusicFX. ImageFX now includes editing controls, allowing users to add, remove, or change specific elements in their images with a simple brush stroke. The tool also incorporates Imagen 3, Google DeepMind's highest quality image generation model, unlocking more photorealism and accurate text rendering.


MusicFX, on the other hand, introduces DJ Mode, a feature that enables users to mix beats by combining genres, instruments, and more. This mode serves as a playground for inspiring new music and pushes the boundaries of AI-powered music creation.



As with all AI technologies, Google Labs is committed to developing VideoFX responsibly. Videos created by Veo are watermarked using SynthID, a cutting-edge tool for identifying AI-generated content. The model also undergoes safety filters and memorization checking processes to mitigate privacy, copyright, and bias risks.

Google Labs is actively collaborating with leading creators and filmmakers to gather feedback and ensure that VideoFX benefits the wider creative community. This ongoing partnership will shape the future of Veo and other generative video technologies.


VideoFX is currently available in private preview starting in the U.S., and interested users can sign up to join the waitlist. ImageFX and MusicFX updates are now accessible in 110 countries and 37 languages, making these powerful tools more widely available to creators worldwide. In the future, Google plans to bring some of Veo’s capabilities to YouTube Shorts and other products.