Microsoft Logo
Credit: Reuters File Photo
Redmond-based technology giant Microsoft has launched the new MAI-Image-1, a generative Artificial Intelligence (AI)-based photo generator.
Unlike the Copilot AI and Office suite of productivity tools and apps, which is powered by OpenAI's ChatGPT, the MAI-Image-1 is an in-house product developed by Microsoft.
Images generated by Microsoft's MAI-Image-1 model.
Credit: Microsoft
"We trained this model with the goal of delivering genuine value for creators, and we put a lot of care into avoiding repetitive or generically-stylized outputs. For example, we prioritised rigorous data selection and nuanced evaluation focused on tasks that closely mirror real-world creative use cases – taking into account feedback from professionals in the creative industries. This model is designed to deliver real flexibility, visual diversity and practical value," said Microsoft.
MAI-Image-1 is a text-to-image generator and is touted to be faster than conventional image generator models available in the market. Users can simply describe a scenario, and it will generate photorealistic images in just a few seconds.
Besides photo-realistic images, it supports several other varieties such as water painting, cyber punk, illustrations, modern colour block style, flat paper cutout-style, claymation style, line art style and many more.
MAI-Image-1 is Microsoft's third fully owned gen AI model after the voice generator MAI-Voice-1 AI and chatbot MAI-1.
MAI-Image-1 will be competing with OpenAI's GPT Image-1, Google's Imagen-4 and Meta AI Imagine, which is powered by the Expressive Media Universe (EMU) model.
Interested users can head to Microsoft's official website (here) to test the new MAI-Image-1- 1 model. There is no official word on when the company plans to bring a standalone mobile app for Android and iOS devices.
Get the latest news on new launches, gadget reviews, apps, cybersecurity, and more on personal technology only on DH Tech.