Meta Launches Muse Image AI Model To Power Image Creation Across Its Apps
The new model is designed to act as a creative assistant, allowing users to generate high-quality visuals using natural language prompts.
Meta has introduced Muse Image, its first image generation model from Meta Superintelligence Labs, expanding the capabilities of Meta AI with advanced image creation and editing features across its platforms.
The new model is designed to act as a creative assistant, allowing users to generate high-quality visuals using natural language prompts. Available through Meta AI, Muse Image enables users to create, edit, and share images directly to Facebook, Instagram, Messenger, WhatsApp, Stories, and chats.
Users can generate images from scratch or modify existing photos by describing the desired changes in conversational language. The model can perform tasks such as removing unwanted objects, placing users in different locations, generating functional QR codes, and creating infographics with cleanly rendered text.
Meta has also introduced more than 30 AI-powered effects for Instagram Stories and image generation within direct chats with Meta AI. To help users get started, the platform includes a presets panel featuring one-tap creative options such as restoring old family photos, trying trending hairstyles, or transforming images into claymation or retro video game-style artwork.
The company is also bringing AI-powered room redesign capabilities to Meta AI. Users can upload a photo of a room and ask the assistant to redesign the space using real products from the web or Facebook Marketplace based on specific styles or current design trends.
According to Meta, Muse Image works alongside its Muse Spark AI model to understand prompts, plan image layouts, retrieve relevant web context, and combine multiple visual references before generating an image. This enables more accurate and context-aware visual creation.
The model also supports Instagram integration, allowing users to mention public Instagram accounts to incorporate publicly available images into personalized visual designs. Meta said users can control whether their public content can be used for AI-generated creations through privacy settings.
Users can further refine generated images by drawing or annotating directly on them, with Meta AI remembering the conversation context for iterative edits.
Meta said Muse Image is currently available through Meta AI and will expand to more countries and additional surfaces across Facebook, Messenger, Instagram, and WhatsApp in the coming months.
The company also plans to make the model available to advertisers through Advantage+ Creative while continuing development of its upcoming Muse Video model.
