MAI-Image-1
Microsoft's first in-house text-to-image model debuts in the top 10 on LMArena; rolled into Bing Image Creator and Copilot.
Added a self-trained image model next to DALL-E 3 and GPT-4o in Bing Image Creator, tuned for photorealism (lighting, reflections, landscapes) with input from creative professionals.
- Date
- Monday, 13 October 2025
- Lab
- Microsoft
- Kind
- model
- Access
- app only
Figures
| Measure | Value | Measured by |
|---|---|---|
| LMArena text-to-image | top 10 at debut exact rank not given in the post | third-party |
Sources
This record was checked against its sources on 6 October 2026. How we check