Microsoft Edge Powers AI Web Apps for Developers
▼ Summary
– Microsoft is launching new AI APIs for Edge, allowing developers to integrate AI functionality into web apps using models built into the browser.
– The APIs, unveiled at Build 2025, compete with Google Chrome’s built-in AI features, aiming to attract developers to Edge.
– Edge’s APIs include access to Phi 4 mini, a small but efficient AI model excelling at math, with 3.8 billion parameters.
– The new APIs also offer writing assistance features and a future translator API, all supporting on-device processing for privacy and security.
– Microsoft is introducing a PDF translation tool in Edge, supporting over 70 languages, with general availability next month.
Microsoft Edge is introducing powerful new AI capabilities for developers, enabling them to build smarter web applications directly within the browser. The company recently unveiled a suite of APIs at Build 2025, giving developers access to built-in AI models like Phi 4 mini, a compact yet capable model optimized for tasks such as mathematical problem-solving.
This move positions Edge as a strong competitor to Google Chrome, which already offers similar AI integration. By embedding these tools directly into the browser, Microsoft aims to attract developers looking for efficient, on-device AI processing without relying on external cloud services.
Phi 4 mini, with its 3.8 billion parameters, strikes a balance between performance and efficiency, making it ideal for devices with limited processing power, such as laptops and smartphones. Unlike larger models, it delivers solid results while minimizing resource consumption.
Beyond Phi 4 mini, Edge is rolling out additional AI-powered APIs, including text generation, summarization, and editing tools. Later this year, Microsoft plans to introduce a translation API, allowing seamless language conversion directly within web apps.
Microsoft emphasizes that these APIs are designed with privacy and security in mind, particularly for developers handling sensitive data. Since processing happens locally, there’s no need to transmit information to third-party servers—a key advantage for regulated industries.
Currently, these features are available in Edge’s Canary and Dev channels, giving early adopters a chance to experiment before wider release.
In a related update, Microsoft is also enhancing Edge’s PDF translation capabilities, supporting over 70 languages. Users can now open a PDF in the browser, click the translate button, and generate a new document in their preferred language. This feature will be fully available next month, with early access starting today for Canary testers.
By integrating AI directly into Edge, Microsoft is making it easier for developers to create cutting-edge web applications while maintaining performance and security—a strategy that could reshape how AI is used in everyday browsing.
(Source: TechCrunch)
