# Microsoft Launches Two AI Models MAI-Image-2.5-Pro and MAI-Voice-2-Flash

- Category: Flash (快讯)
- Published: 2026-07-24T11:56:00.000Z
- Source date: 2026-07-24
- HTML: https://www.88z.ai/en/news/flash-20260724-81e9d4834c97f48f
- Markdown: https://www.88z.ai/en/news/flash-20260724-81e9d4834c97f48f.md
- Other language: https://www.88z.ai/news/flash-20260724-81e9d4834c97f48f.md

Microsoft has officially released two self-developed AI models, MAI-Image-2.5-Pro and MAI-Voice-2-Flash, opening them for public preview. The former is the company's most accurate image generation model to date, supporting main visual creation, detail editing, and text rendering. The latter is optimized for high-concurrency voice scenarios and supports 15 languages.

## At a glance

- MAI-Image-2.5-Pro is Microsoft's most accurate image model, supporting visual generation and text rendering
- MAI-Voice-2-Flash is optimized for high-concurrency scenarios and supports 15 languages

## The update

Microsoft has officially launched two self-developed artificial intelligence models built on internal training processes: MAI-Image-2.5-Pro and MAI-Voice-2-Flash, both of which are now available for public preview. MAI-Image-2.5-Pro represents the highest level of accuracy achieved by Microsoft in image generation to date. This model not only possesses powerful capabilities for generating main visuals but also supports fine-grained editing of image details and high-quality text rendering within images. Meanwhile, MAI-Voice-2-Flash is specifically optimized for high-concurrency voice processing scenarios, aiming to enhance response efficiency and stability in complex environments while broadly supporting voice interaction needs in 15 different languages. The release of these two models marks a further expansion of Microsoft's layout in multimodal AI technology.

## Sources

- [IT之家](https://www.ithome.com/0/980/899.htm) (2026-07-24)

This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.