Superseded by newer MAI models, so mostly of interest as a free image option inside Bing Image Creator.

Kenning's take

What it does

MAI-Image-1 was the first text-to-image model Microsoft AI built entirely in-house, announced in October 2025 with a focus on photorealistic lighting, outdoor scenes and fast output. From November 2025 it appeared as a model choice in Bing Image Creator next to DALL-E 3 and GPT-4o, and powered images in Copilot Audio Expressions story mode. Microsoft has since released MAI-Image-2 and MAI-Image-2.6, so it is no longer the newest MAI image model.

  • Text-to-image generation
  • Photorealistic lighting and landscapes
  • Model option in Bing Image Creator
  • Images for Copilot Audio Expressions story mode

Who it suits. Bing and Copilot users making images from text prompts.

Key facts

Category
AI models
Made by
Microsoft AI
Based in
United States
Released
13 October 2025
How to get it
launched in Bing Image Creator (bing.com/create and the Bing app) and Copilot Audio Expressions; current availability alongside newer MAI models not confirmed on the pages fetched
Works on
Web, iOS, Android
Free plan
Free in Bing Image Creator at launch

FAQ

When was MAI-Image-1 released?

Microsoft announced it on 13 October 2025 and began adding it to Bing Image Creator on 4 November 2025.

Where could I use MAI-Image-1?

In Bing Image Creator at bing.com/create, in the Bing app, and in Copilot Audio Expressions story mode.

Is MAI-Image-1 Microsoft's latest image model?

No. Microsoft has since released MAI-Image-2 and MAI-Image-2.6.

Facts from the maker's own pages and the sources listed. Checked 2026-10-11.