Meta Platforms (META) is making an aggressive push into the generative AI arena with the launch of Muse Image, a new artificial intelligence model designed to create and edit pictures from text prompts. The rollout, announced on Tuesday, marks the first image-generation model from the company’s recently established Meta Superintelligence Labs and is being woven directly into the fabric of its social media empire, including WhatsApp and Instagram.
Muse Image is engineered to interpret complex user commands, accept photos as reference inputs, and allow for granular edits through sketches or annotations. The tool is integrated directly into the Meta AI chatbot, signaling the company’s intent to transform its assistant into a full-fledged creative suite rather than a simple text-based query tool. The immediate application is massive: Meta stated that Muse Image will power more than 30 new AI effects for Instagram Stories and enable image generation inside direct chats with Meta AI on WhatsApp, starting in select markets.
“Muse Image brings native reasoning to the creative process to adjust elements, swap styles, and create variations,” the company explained in a business-focused blog post, highlighting the model’s potential to streamline marketing workflows.
This launch is the second major release from Meta Superintelligence Labs, a specialized team assembled last year under the leadership of Alexandr Wang to close the gap with rivals like OpenAI and Google. The unit debuted its first model, a text-and-reasoning large language model called Muse Spark, in April, which replaced the legacy Llama family of models. The rapid succession of Muse Spark and now Muse Image—originally codenamed Mango—demonstrates the urgency with which CEO Mark Zuckerberg is pivoting resources toward monetizing AI.
Meta is adopting a freemium model to drive adoption while building a new subscription revenue stream. Basic access to Muse Image via the Meta AI app, website, WhatsApp, and Instagram Stories will be free, but usage is capped. To generate a high volume of AI images or unlock advanced editing features, power users and creators must subscribe to the Meta One monthly plan, which debuted in May. If users hit their free limit, they face a choice: pay for a subscription or wait for the counter to reset. This strategy mirrors moves by competitors who are attempting to offset the massive computational costs of generative AI.
Beyond consumer subscriptions, Meta is targeting its core advertising business. Muse Image will be integrated into the AI-powered Advantage Plus service, an automated toolkit that helps brands develop ad creative and manage campaigns. By powering advertiser-specific image generation, Meta aims to make it easier for businesses to produce high-quality, on-brand variations with fewer manual iterations. The company confirmed it has been collaborating with businesses and advertisers during the development phase to ensure the tool meets commercial needs.
The battle for AI image generation supremacy is intensifying. While Meta is a dominant force in social media, it has been perceived as a laggard in generative AI compared to OpenAI and Alphabet (GOOGL). Google’s Nano Banana model became a consumer sensation upon its release last fall, giving the search giant an early lead in public mindshare. Meta is now openly benchmarking its technology against these rivals. Internal tests revealed that Muse Image trails OpenAI’s latest GPT Image 2 model in complex editing tasks but outperforms the Nano Banana 2 model in scenarios involving the editing of both single and multiple images.
Historically, Meta relied on third-party AI models, such as Midjourney and Black Forest Labs, to power visual generation features within its ecosystem. The launch of Muse Image represents a strategic move toward vertical integration, reducing dependency on external technology providers and giving Meta full control over the user experience and cost structure. The company also teased an early preview of Muse Video, a forthcoming video generation model that it claims offers “competitive performance in prompt adherence, visual fidelity, and temporal consistency.”
The rollout will be phased. While the initial launch focuses on Instagram and WhatsApp, Meta plans to expand Muse Image to Facebook and Messenger, as well as deeper into Instagram and WhatsApp, later in the year. This gradual integration suggests a cautious approach to scaling, ensuring that the computational load and content moderation challenges do not outpace the infrastructure.
For investors, the Muse Image launch is a tangible signal of how Meta intends to monetize its hefty AI infrastructure spending. By embedding generative AI directly into the advertising pipeline through Advantage Plus, Meta is not just chasing consumer trends but seeking to increase the efficiency and spending of the millions of businesses that advertise on its platforms. The dual revenue approach—combining consumer subscriptions with enhanced ad tools—provides a hedge against the volatile digital advertising market while potentially increasing user lock-in across its family of apps.