Microsoft Unveils Seven New AI Models with Enhanced Capabilities in Reasoning, Coding, Imaging, and Voice Processing

Microsoft Unveils Seven New AI Models at Annual Build Conference

Microsoft has made a significant push into artificial intelligence with the unveiling of seven new AI models at its annual Build developer conference. This marks one of the company's largest AI product launches to date and highlights its growing efforts to develop in-house AI technology.

The new models include MAI-Thinking-1, MAI-Code-1-Flash, MAI-Image-2.5, MAI-Image-2.5-Flash, MAI-Transcribe-1.5, MAI-Voice-2, and MAI-Voice-2-Flash. At the center of the announcement is MAI-Thinking-1, Microsoft's first advanced reasoning model. According to the company, MAI-Thinking-1 was trained from the ground up using commercially licensed data and was not distilled from third-party frontier models. Microsoft describes it as a medium-sized model designed for complex reasoning, coding, and long-context tasks while maintaining lower operating costs.

The launch reflects Microsoft's strategy to strengthen its own AI capabilities alongside its partnership with OpenAI. Industry analysts view the move as an effort by the technology giant to gain greater control over AI development, costs, and product differentiation as competition intensifies across the sector.

Read also: MNC IPO Activity to Remain a Key Theme in Indian Market, Citi Analyst Predicts

In addition to MAI-Thinking-1, Microsoft also introduced MAI-Code-1-Flash, a lightweight coding model built for GitHub Copilot and Visual Studio Code. This model is designed to deliver strong coding performance while using fewer computing resources, making it more efficient for everyday software development tasks.

The company also introduced MAI-Image-2.5 and its Flash variant for text to image generation and image editing. Microsoft said the models offer significant improvements in image quality, visual reasoning, and text rendering, and are being integrated into products such as PowerPoint and OneDrive.

Microsoft also introduced MAI-Transcribe-1.5, a new speech-to-text model that offers faster transcription and improved accuracy. Alongside it, the company unveiled the MAI-Voice-2 lineup, which supports more languages and provides a wider range of voice options for developers creating AI-powered voice assistants and other conversational applications.

The new models will be integrated across Microsoft's products, developer tools, and cloud services. This launch expands the company's portfolio of in-house AI models and adds new capabilities in reasoning, coding, image generation, speech recognition, and voice applications.

Read also: India Seeks Resolution of ONGC Dues in Talks with Venezuela