Microsoft’s first internal AI image generator, MAI-Image-1, is now available in two products: Bing Image Creator and Copilot Audio Expressions. The company announced the model in October. Microsoft’s AI chief Mustafa Suleyman wrote in: write to X that the text-to-image model will appear “soon” in the EU.
Suleyman added that the model “really excels” in creating images of food and nature, as well as artistic lighting and photorealistic details.
Microsoft has previously published more details about this solution your blog: “MAI-Image-1 specializes in generating photorealistic images such as lighting (e.g. reflected light, reflections), landscapes and much more. This is especially true when compared to many larger, slower models. The combination of speed and quality means users can get their ideas on screen faster, review them quickly, and then transfer their work to other tools for further refinement.”
Microsoft’s MAI-Image-1 will also create AI-generated graphics to accompany AI-generated audio stories in the “Story Mode” of Copilot’s text-to-speech platform, Copilot Audio Expressions.
MAI-Image-1 is listed as one of three artificial intelligence models available on the Bing imaging website and app. The other two models, DALL-E 3 and GPT-4o, come from OpenAI.