Midjourney develops 'Patchwork,' a giant canvas tool that combines image generation AI and text generation AI to create

Midjourney, a company that provides image generation AI, has announced the release of Patchwork, a multiplayer world-building tool. Patchwork combines language models and image generation models to allow multiple people to collaborate on creating a single world.

Patchwork Research Preview

Today we're opening up a new experimental multiplayer worldbuilding tool called 'Patchwork'. It combines language models, image models, and a canvas-based interface to build out the foundations of stories. Check out the ????below for links and further documentation. Have fun! pic.twitter.com/IgoJrJ8bO9

— Midjourney (@midjourney) December 11, 2024

Patchwork allows you to develop vague story ideas into something more concrete, collaborate with others to create a coherent setting, and even assemble fragments of images and text into collage-like visual stories.

Patchwork is based on an image generation AI that can generate images from text. You can then choose your favorite style from the generated images and select the 'world' you want to create.

The shared board looks like this. You can post the generated image or enter text on the spot to generate a new image.

Multiple people can access one board, and by sharing the images each person creates on the shared board, it becomes possible to align each member's image.

In the future, Midjourney plans to make the characters and worlds created in Patchwork available to other storytelling apps, allowing users to move characters around in interactive stories, guide text generation by acting out story scenes with gestures, and even use new LLM-based interfaces for creative writing.

At the time of writing, Patchwork is available as a research preview and can be tried out at patchwork.midjourney.com . A Midjourney account and Google login are required to use it. However, Midjourney says that Patchwork is still in development and may have some issues.