Microsoft has officially launched a new AI image generator, the MAI-Image-1, its first fully in-house developed image generation model. The model has already made a strong debut, ranking in the top 10 on LMArena’s text-to-image benchmarks.
As of earlier this week, MAI-Image-1 has begun rolling out across select Microsoft platforms:
Bing Image Creator: Users can now access MAI-Image-1 at bing.com/create, through the Bing mobile app, or directly from the Bing search bar. The model appears alongside existing options like DALL-E 3 and GPT-4o, giving creators the flexibility to choose the tool that best fits their project needs.
Copilot Audio Expressions: When using Story Mode, the feature now leverages MAI-Image-1 to generate unique visuals that bring audio stories to life.
The service is currently available in all regions with Bing Image Creator and Copilot Labs access, though some availability is pending.
Built for Real-World Creators
What makes MAI-Image-1 interesting is the deliberate approach to development. Rather than chasing generic outputs, the team focused on delivering practical value for professional creators through:
Rigorous data selection to avoid repetitive or overly-stylized results
Real-world evaluation metrics informed by feedback from creative industry professionals
Visual diversity that offers genuine flexibility for different creative applications
Technical Strengths
MAI-Image-1 particularly excels in photorealistic rendering, with notable capabilities in:
- Advanced lighting effects, including bounce light and reflections
- Detailed landscape generation
- Speed-optimized performance that rivals larger, slower models
This combination of speed and quality enables a more efficient creative workflow—users can rapidly visualize ideas, iterate quickly, and seamlessly export their work to other tools for further refinement.
MAI-Image-1 represents the latest step in Microsoft’s broader mission to create “AI for everyone.” Following the announcement of their first two in-house models in August 2025, this release demonstrates Microsoft’s commitment to developing purpose-built AI solutions rather than relying solely on third-party technologies.
By bringing image generation capabilities in-house, Microsoft gains greater control over model performance, safety features, and integration with its existing product ecosystem—a strategic advantage in an increasingly competitive AI landscape.
To try MAI-Image-1, it’s now available through Bing Image Creator and Copilot Audio Expressions, offering a glimpse into Microsoft’s evolving vision for accessible, practical AI tools.
Leave a Reply