Meta said Muse Image can generate images from text descriptions or modify existing photos. Users can request edits such as removing unwanted background objects, creating visuals with readable embedded text, generating QR codes, or placing themselves into new scenes. The model also supports combining multiple images into a single creation.
The company said Muse Image performs several reasoning steps before producing an image, including planning layouts, using real-time web context, and blending multiple visual references. Meta says this process allows the system to better understand complex prompts and produce results that more closely match a user’s request.
To simplify image creation, Meta AI now includes a presets panel with suggested prompts. Users can restore old family photos, experiment with different hairstyles, or transform themselves into styles such as claymation characters or 16-bit video game heroes. Another feature allows users to photograph a room and ask Meta AI to redesign it using products from the web or Facebook Marketplace.
Muse Image also introduces new editing controls that let users sketch, circle, or annotate parts of an image to indicate changes. Because Meta AI retains the context of the conversation, users can continue refining an image through follow-up prompts without starting over.
The model also integrates with Instagram in a new way. Users can @-mention Instagram accounts inside the Meta AI app to incorporate public photos from those profiles into generated images. Meta said people can disable this capability through a setting that controls whether their content can be used in AI creations.
Beyond consumer features, Meta is extending Muse Image to advertisers through Advantage+ creative. The company said businesses will be able to generate image variations for advertising campaigns using the new model.
“Muse Image brings native reasoning to the creative process to adjust elements, swap styles, and create variations based on the advertiser’s creative, resulting in high-quality, on-brand ad variations with fewer iterations,” Meta said in a blog post aimed at businesses. “In the coming weeks, advertisers and agencies can expect to see image variants powered by Muse Image.”
According to Meta, Muse Image will be free for everyday image generation, while higher usage limits and certain features are available through the company’s subscription plans.
Meta also confirmed that Muse Video is in development. The upcoming model is intended to expand the company’s AI media generation capabilities beyond images and into video creation.
This analysis is based on reporting from Meta.
Image courtesy of Meta.
This article was generated with AI assistance and reviewed for accuracy and quality.
