Google redesigns Flow for AI content creation

Google has redesigned Flow so you can create images and videos, organize them, and edit them in one place. The platform is built for projects such as short films, music videos, and product campaigns, and has already generated more than 1,500 million images and videos since its launch last year.

The update changes Flow’s role: it is no longer focused only on generating videos and now works as a complete creative workspace. Image generation takes center stage, with tools for creating, combining, and modifying assets before turning them into animated scenes.

Images and videos in the same workflow

Google is integrating several capabilities from Whisk and ImageFX, its experimental image-generation tools, into Flow. It is also bringing Nano Banana into the main experience, so you can generate an image and use it directly as a reference or frame for creating a video with Veo, Google’s video-generation model.

In practice, you can create an image of a character, setting, or product and then ask Flow to bring it to life. You can also combine multiple images and style references to maintain a consistent look across different scenes.

Image generation will be available for free inside Flow. In addition, starting in March, early users will be able to transfer their Whisk and ImageFX projects and assets to the Flow library.

A library to keep your work organized

Creating content often means dealing with many versions, tests, and scattered assets. To make that part of the process easier, Flow is introducing an asset grid that lets you:

• Search, filter, and sort images and videos.
• Group related files into collections.
• Switch between an overview and a detail-focused view.
• Drag assets to reorganize them or use them in a prompt.
• Find and reference specific elements with the @ symbol.

This is especially useful when you are working with multiple versions of a character, product, or scene and need to return to one of them without searching through the entire library.

More precise editing with simple instructions

Flow also adds controls for correcting an image without starting over. With the new freeform selection tool, you can mark a specific area and write what you want to change, such as “remove the man” or “add koi fish to the water.” You can also draw directly on the image to indicate the area that should be modified.

For videos, you can now:

• Generate what happens after a clip to extend the scene.
• Add or remove objects with text instructions.
• Remove unwanted people or objects.
• Specify camera movements, such as pans and zooms.

The important change is control. The first version you generate no longer has to be the final result: you can correct details, adjust the composition, and continue developing the same story within the project.

Flow is moving closer to a complete production tool, rather than just a clip generator. What to watch now is how project migration from Whisk and ImageFX will work, and how well these controls can maintain consistent characters, styles, and settings in longer videos.