Microsoft is rolling out a suite of new AI-powered APIs for its Edge browser, giving web developers tools to build smarter applications directly within the browser environment.

The announcement came Monday during the company’s Build 2025 developer conference.

The new features enable websites and Edge extensions to harness Phi-4 Mini, a compact yet powerful AI model with 3.8 billion parameters. According to Microsoft, Phi-4 Mini excels at tasks like solving math problems thanks to a training blend of human-written and AI-generated data.

The APIs are designed to offer developers access to AI functionality, such as text generation, summarization, and editing, without relying on cloud-based processing. Instead, the models run locally on users’ devices, helping ensure privacy and compliance with data regulations.

“These experimental APIs are intended as potential web standards and will work across platforms, browsers, and with other AI models,” Microsoft said in a statement. “For developers dealing with sensitive data or working in regulated industries, these APIs offer the privacy and security of on-device processing.”

Microsoft also revealed plans to launch a translator API in the coming months, enabling AI-powered real-time language translation. The tools are now available for testing via the Edge Canary and Dev channels.

The update positions Edge to compete more directly with Google Chrome, which already offers AI hooks for developers. By integrating AI models into the browser itself, Microsoft aims to reduce friction for developers building advanced web applications.

In a related move, Edge is also gaining a PDF translation tool that supports over 70 languages. Users can translate documents directly in the browser by clicking the “Translate” icon in the address bar. The feature is available today for Canary users and will roll out broadly next month.