Microsoft Unveils New APIs to Bring On-Device AI to Web Apps in Edge Browser
Microsoft is expanding the capabilities of its Edge browser to include on-device artificial intelligence (AI) models, allowing web developers to leverage these powerful tools directly within their applications. The announcement was made during Microsoft's Build conference, held today. Developers will gain access to new APIs that can integrate Microsoft’s Phi-4-mini model, a compact AI model with 3.8 billion parameters launched in February. The Phi-4-mini model, along with its larger counterpart Phi-4, offers advanced features such as text generation, summarization, and editing. By integrating these APIs, web developers can enhance their applications with interactive elements like prompt boxes, providing users with real-time assistance and improving overall user experience. Additionally, Microsoft plans to release a text translation API within the next few months, further broadening the scope of AI-driven functionalities available. These APIs are not only cross-platform, supporting both Windows and macOS, but they are also proposed as potential web standards. This means that they could eventually be adopted by other browsers, fostering a more consistent and ubiquitous AI-powered web environment. Developers interested in testing these new APIs can do so now through the Edge Canary and Dev channels. Google, a major competitor in the browser market, already offers similar APIs for its Chrome browser. These APIs enable developers to utilize Chrome’s built-in AI models for tasks such as text translation, text and image generation, and even creating calendar events based on webpage content. This indicates a growing trend among tech giants to integrate AI capabilities into their browsers, aiming to provide richer and more dynamic user experiences. By making these on-device AI models accessible to web developers, Microsoft is positioning Edge to be a leading platform for innovative web applications. The move underscores the company’s commitment to advancing AI technology and making it more widely available to users and developers alike. As these APIs mature and potentially become standardized, they could significantly influence the future of web development and user interaction.
