Microsoft announced the launch of MAI-Image-2, a new text-to-image model developed by its AI Superintelligence team, marking a significant step in the company’s push to build its own in-house AI capabilities. The model positions Microsoft among the top three AI image-generation labs globally, according to the Arena.ai leaderboard.
MAI-Image-2 is designed to enhance creative workflows by focusing on photorealism, accurate text rendering within images, and the ability to generate complex and imaginative scenes. The company said the model was developed in collaboration with photographers, designers, and visual storytellers to better align with real-world creative needs.
The model emphasizes improved realism, including natural lighting, accurate skin tones, and environments that feel more lifelike. Microsoft also highlighted advancements in generating text within images, enabling use cases such as posters, infographics, and visual presentations with greater consistency. Additionally, MAI-Image-2 is built to handle detailed and surreal compositions, supporting more ambitious and cinematic visual concepts.
MAI-Image-2 is currently available through the MAI Playground, where users can experiment with the model and provide feedback. Microsoft noted that the model is beginning to roll out across its ecosystem, including Copilot and Bing Image Creator. API access is already available to select enterprise customers, with broader availability expected through Microsoft Foundry.
The launch reflects Microsoft’s broader strategy to reduce reliance on external AI providers by developing competitive models internally. By building its own image-generation capabilities, the company aims to gain greater control over performance, cost, and product integration across its platforms.
Microsoft indicated that additional updates and capabilities are expected as the MAI-Image-2 model continues to evolve, positioning it as a foundational component of its future AI offerings.
