Meta has launched Muse Image, the first image-generation model created by its Superintelligence Labs, and it’s now available in Meta AI. The release builds on Muse Spark, which previously enhanced the assistant’s reasoning capabilities. Muse Image is designed to help users create high-quality images from their ideas, with options to download or share them directly in chats, stories, or social feeds.
The model also powers several new creative features across Meta’s apps. Instagram Stories gains more than 30 AI-powered effects, while WhatsApp users can generate images directly in one-on-one chats with Meta AI. The rollout begins in select countries before expanding to additional markets.
Muse Image understands natural language, making it easy to create new images or edit existing ones. Users can ask it to place themselves in front of a famous landmark, remove someone from the background of a photo, or generate a functional QR code based on a custom prompt. The model also handles text inside images well, making it suitable for creating how-to graphics, infographics, and other designs with clear, readable typography.
The model is built to handle more complex requests, using advanced reasoning to interpret detailed prompts and combine multiple images into a single result. It also includes suggested prompts for inspiration, an @mention feature for adding photos to an existing project, and drawing tools that let users mark up an image to guide edits.
Meta plans to expand the rollout of Muse Image in the months ahead. The company says the model will become available in more countries and across additional Meta AI experiences, including Facebook, Messenger, Instagram, and WhatsApp. Advertisers and agencies are also expected to gain access through Advantage+ Creative in the coming weeks.
Maybe you’d like some other interesting articles?