Microsoft announced a fresh AI system on Thursday through a post on its AI blog. Named MAI-Image-2, this tool focuses on creating visuals from textual descriptions, featuring unique enhancements.
The company states that MAI-Image-2 excels at producing realistic photos incorporating aspects such as realistic illumination, precise complexion representations, and immersive settings. Designed specifically for artistic applications, Microsoft notes that input from photographers, graphic artists, and narrative visual experts guided its development to address key needs in routine creative tasks.
Beyond lifelike visuals, MAI-Image-2 reliably incorporates lettering into its outputs, positioning it as an ideal choice for elements like charts, presentations, charts, and similar materials. It also handles imaginative designs, detailed arrangements, and expansive fictional realms effectively.
Thanks to MAI-Image-2, Microsoft's MAI system has advanced in performance, now holding the third position among text-to-image developers on the Arena.ai ranking.
The model is being deployed progressively on Copilot and Bing Image Creator platforms. Users can test the newest MAI versions through the MAI Playground, where they can also submit comments to the development team.
This piece first appeared in our affiliated outlet PC för Alla and underwent adaptation and translation from its Swedish version.