Microsoft AI has launched MAI-Image-2.5, a text-to-image model ranked third on the Arena leaderboard.
Microsoft AI has launched MAI-Image-2.5, its latest text-to-image model, with the model ranking third on the Arena text-to-image leaderboard.
The model is part of the MAI-Image series and is available to try on Arena. Microsoft AI says in a recent blog post it will also come to the MAI Playground and Microsoft Foundry within the next two weeks.
Mustafa Suleyman, CEO of Microsoft AI, announced the launch ahead of Build on LinkedIn, saying the model delivers improvements in text rendering, cartoon generation, and commercial imagery.
MAI-Image-2.5 follows MAI-Image-1 and MAI-Image-2, with Microsoft AI positioning the new release around more reliable outputs for creative and brand-focused work.
For education, learning design, and workforce skills teams using generative AI, the launch adds another model to the fast-moving image generation market, where text accuracy, layout control, and consistent visual composition are becoming practical requirements rather than nice-to-have features.
MAI-Image-2.5 is ranked third on the Arena text-to-image leaderboard, according to Microsoft AI. The company says the model performs across a range of styles, follows instructions closely, renders text more reliably, and produces detailed and coherent images.
Suleyman said on LinkedIn: “It’s our strongest image model yet and a real step change in quality, delivering major improvements in text rendering, cartoon generation and commercial imagery.”
He added: “MAI-Image-2.5 performs where you need it most. Words are sharper. Layouts hold together better. Scenes feel more deliberate. Brand-forward visuals come through with more polish.”
Microsoft AI says the model shows stronger visual reasoning across objects, scene structure, lighting, scale, and spatial relationships. Those capabilities are relevant for product shots, posters, packaging concepts, learning visuals, training assets, and marketing materials where prompt adherence and visual structure affect whether an image can be used.
The blog post says: “Professional grade creative work requires getting every detail right: the words on a poster, the label on packaging, the structure of a product shot, the way light falls across a scene.”
Text rendering has been one of the recurring weak points in AI image generation, particularly for use cases involving posters, product labels, branded templates, classroom visuals, worksheets, diagrams, and campaign materials.
Microsoft AI says MAI-Image-2.5 makes gains in text rendering and product and branding concepts. The announcement points to sharper words, better layout structure, more deliberate scenes, and more polished brand-forward visuals.
The model also improves stylized illustration and commercial imagery compared with MAI-Image-2, according to Microsoft AI.
The launch comes as image models are increasingly being tested not just on visual quality, but on whether they can follow detailed creative instructions and produce outputs with usable text, consistent scale, and coherent layouts.
MAI-Image-2.5 is available on Arena now. Microsoft AI says the model is also coming to the MAI Playground and Microsoft Foundry within the next two weeks.
Suleyman said: “With Build round the corner, there’s so much more to come from the Microsoft AI team.”
Microsoft AI has not disclosed pricing, licensing terms, enterprise access details, education-specific availability, or whether MAI-Image-2.5 will be integrated into other Microsoft products. The next confirmed rollout points are MAI Playground and Microsoft Foundry.
Sign up with your email address to receive news and updates.
We respect your privacy and will not pass your email address on to third parties. However, we will occasionally send you promotional messages on behalf of our advertisers.
© EdTech Innovation Hub. All rights reserved. No part of this website, including text, graphics, logos, images, and software, may be reproduced, distributed, or transmitted in any form or by any means without the prior written permission of EdTech Innovation Hub.
Leave a Reply