We've added 18 new Style References that have been optimized for GPT Image.

Generate images with OpenAI’s new state-of-the-art image model right from Visual Electric.

You can now animate any image (including personal uploads and AI generations) right from your canvas. Just select your image and tap “Video” from the sidebar.

“Styles” have been renamed to “References”. All References are now shared with your workspace — and we’re just getting started. Image, character, and object References are on the way.

You can now train a custom model on your own data. Just import your own images to a new Reference, and we’ll handle the rest. Accessible to your whole workspace in ~5