Microsoft introduces its first text-to-image model and promises better pictures
After releasing its own artificial intelligence models, Microsoft has announced another AI-powered tool: its first text-to-image model.
Image by Cybernews
After releasing its own artificial intelligence models, Microsoft has announced another AI-powered tool: its first text-to-image model.
The company says that its image generation model, MAI-Image-1, is already among the top 10 text-to-image models on LMArena, a platform that evaluates large language models.
Microsoft shared that the aim of this model is to provide users with valuable outputs, “avoiding repetitive or generically-styled” production. To achieve that, the company gathered feedback from industry professionals, prioritizing data that provides insights into “real-world creative use cases.”
“This model is designed to deliver real flexibility, visual diversity and practical value,” shared Microsoft in the post.
According to the company, MAI-Image-1 stands out from other AI models by generating more photorealistic images that include realistic details such as reflections, and doing it faster.
The new AI image generator is now available through LMArena, with extended availability planned for later on Copilot and Bing Image Creator.
A few months ago, Microsoft announced its two in-house built artificial intelligence models, MAI-Voice 1 and MAI-1-preview.
With MAI-Voice 1, users can generate text in a specific style, tone, and voice through textual prompts, while MAI-1-preview can be used as a chatbot that could answer various questions.
Unlock more exclusive Cybernews content on YouTube.