MAI-Image-2.5

MAI-Image-2.5 is Microsoft AI’s updated flagship image-generation model, purpose-built for high-quality text-to-image generation and precise, controllable image-to-image editing at production scale. It introduces a suite of “control with preservation” capabilities — identity and character consistency across stylization, pose, and layout; localized edits that leave the rest of the image untouched; structured document and diagram generation that produces PowerPoint-ready visuals and slides. The model understands scene structure, lighting, scale, and spatial relationships, and ships in both the standard MAI-Image-2.5 and a faster MAI-Image-2.5-Flash variant for high-volume workloads.

On the Arena.ai community leaderboards, MAI-Image-2.5 places among the top three model families on both text-to-image and image-editing — debuting at #3 with an average +74.5 ELO improvement over MAI-Image-2, including +104 on text rendering and +90 on cartoon, anime, and fantasy. The combination of editing capability, identity preservation, and slide-ready output makes it the engine for enterprise creative and productivity workflows — designers iterating on a single subject across compositions, marketers stylizing assets at scale, and Office users generating presentation visuals without leaving the application.