Meta has introduced Muse Image, a new artificial intelligence image-generation model that enables users to create high-quality visuals from simple text prompts and ideas.
Developed by Meta Superintelligence Labs, the tool is the company’s first dedicated image-generation model and has been integrated into Meta AI, where it can be used to create, edit and share images across Meta’s family of apps.
The company said Muse Image combines advanced reasoning with image creation, allowing it to interpret detailed instructions, understand context and merge multiple visual references into realistic, personalised images.
According to Meta, the new model expands on the capabilities of Muse Spark, its large language model, by extending AI intelligence beyond text generation into visual content creation.
In a statement issued on Wednesday, Meta said the feature enables users to transform ideas into polished visuals that can be downloaded or shared directly to feeds, stories and chats.
The company noted that Muse Image can also handle specialised requests, including placing users in different environments, redesigning existing images, producing infographics and creating customised QR codes.
To make image creation easier, Meta has introduced ready-made creative presets that allow users to restore old family photographs, experiment with artistic styles and convert ordinary images into unique digital artwork.
Beyond creative use, the company said the technology offers practical applications such as interior design.
Users can upload pictures of their homes and ask Meta AI to redesign rooms using preferred styles while suggesting products available online or through Facebook Marketplace.
Meta explained that Muse Image analyses each prompt before generating an image, enabling it to plan layouts, combine several references and produce results that more accurately reflect users’ intentions.
The company also announced a feature that lets users tag Instagram accounts within the Meta AI app so publicly available images from those profiles can be incorporated into AI-generated creations. It added that users remain in control of whether their content is available for AI image generation through privacy settings.
Meta further said editing has become more intuitive, with users able to modify generated images by sketching directly on them, highlighting areas for changes or requesting refinements without having to start the process from scratch.
