Microsoft has introduced MAI-Image-1, its first text-to-image generation model built entirely in-house. The company says the model focuses on delivering photorealistic images with an ideal balance between speed and quality. Currently available for public testing on LMArena, MAI-Image-1 will soon be integrated into Microsoft’s Copilot and Bing Image Creator platforms, which currently use OpenAI’s GPT-4o and DALL-E 3.
This marks Microsoft’s third homegrown AI model, following the launch of MAI-Voice-1 and MAI-1-preview in August. The company aims to strengthen its independent AI capabilities to power Copilot and other creative tools.
According to Microsoft, MAI-Image-1 ranks among the top 10 text-to-image models on LMArena, joining other leading models such as Google Gemini 2.5 Flash Image and OpenAI’s GPT-1. It is positioned to compete with Google’s Imagen 3 and similar models in the text-to-image space.
Model development and features
Microsoft stated that MAI-Image-1 was trained with a strong focus on data selection and evaluation suited to real-world creative tasks. The company worked with professionals in creative industries to refine the model and reduce repetitive or oversimplified visual styles.
MAI-Image-1 emphasises photorealism, showcasing its ability to handle complex lighting, reflections, and landscapes effectively. Microsoft described it as a fast and efficient model that allows creators to bring concepts to life quickly while maintaining high visual quality. The model aims to support creative workflows by offering visual diversity and enabling smooth integration with other design tools.
Safety and deployment
Microsoft said safety and responsibility are key priorities in the model’s development. The company is testing MAI-Image-1 on LMArena to gather feedback from users before full deployment. Microsoft confirmed that the model will be added to Copilot and Bing Image Creator “very soon.”
Other MAI models
MAI-Voice-1 is Microsoft’s first expressive speech generation system that can produce one minute of audio in under a second using a single GPU. It supports both single- and multi-speaker scenarios with high-fidelity output.
MAI-1-preview is a mixture-of-experts foundation model trained on approximately 15,000 NVIDIA H100 GPUs. It is designed for text-based instruction following and is also available for testing on LMArena.
Also read: Viksit Workforce for a Viksit Bharat
Do Follow: The Mainstream formerly known as CIO News LinkedIn Account | The Mainstream formerly known as CIO News Facebook | The Mainstream formerly known as CIO News Youtube | The Mainstream formerly known as CIO News Twitter |The Mainstream formerly known as CIO News Whatsapp Channel | The Mainstream formerly known as CIO News Instagram
About us:
The Mainstream formerly known as CIO News is a premier platform dedicated to delivering latest news, updates, and insights from the tech industry. With its strong foundation of intellectual property and thought leadership, the platform is well-positioned to stay ahead of the curve and lead conversations about how technology shapes our world. From its early days as CIO News to its rebranding as The Mainstream on November 28, 2024, it has been expanding its global reach, targeting key markets in the Middle East & Africa, ASEAN, the USA, and the UK. The Mainstream is a vision to put technology at the center of every conversation, inspiring professionals and organizations to embrace the future of tech.
