Microsoft Unveils On-Device AI APIs for Edge Browser, Empowering Web Developers

Devs can now tap Microsoft Edge to power AI web apps | TechCrunch

Microsoft is launching new APIs for Edge, its web browser, to let developers incorporate AI functionality into web apps using models built into Edge. Unveiled at Build 2025, the AI APIs mirror some of the functionality in Google Chrome, which also offers "built-in AI" that developers can tap to power their web applications. Microsoft's move suggests that it's keen to offer a competitor -- one that might sway devs to its side. Edge's new APIs give websites and Edge browser extensions access to Phi 4 mini, an AI model that Microsoft released in late February. Around 3.8 billion parameters in size, Phi 4 mini excels at math problems, Microsoft says, thanks to a training data set that includes a mix of human-crafted and synthetic data (i.e. data generated by AI models). Parameters roughly correspond to a model's problem-solving skills, and models with more parameters generally perform better than those with fewer parameters. Small models have the advantage of being more efficient, however, particularly on less powerful hardware like that found in laptops and mobile devices. Edge is also gaining other AI APIs, including a set of writing assistance APIs for generating, summarizing, and editing text. In a few months, Microsoft plans to release a translator API to serve AI-powered language translations via Edge, the company says. "These experimental APIs are intended as potential web standards and will work across platforms, browsers, and with other AI models," writes Microsoft in a press release provided to TechCrunch. "For developers dealing with sensitive data or working in regulated industries, these APIs offer the privacy and security of on-device processing, eliminating the need to send data to external cloud services." All of the new AI APIs are available in the Edge Canary and Dev channels. In related news, Microsoft is bringing a translation tool for PDFs to Edge. With support for over 70 languages, the tool lets users open a PDF in Edge, click the "Translate" icon in the Edge address bar, and create a new doc translated into the language of their choice. PDF translation will be generally available next month, and Canary users can try it starting today.

Microsoft is opening its on-device AI models up to web apps in Edge

Wes Davis is a weekend editor who covers the latest in tech and entertainment. He has written news, reviews, and more as a tech journalist since 2020. Web developers will be able to start leveraging on-device AI in Microsoft's Edge browser soon, using new APIs that can give their web apps access to Microsoft's Phi-4-mini model, the company announced at its Build conference today. And Microsoft says the API will be cross-platform, so it sounds like these APIs will work with the Edge browser in macOS, as well. The 3.8-billion-parameter Phi-4-mini is Microsoft's latest small, on-device model, rolled out in February alongside the company's larger Phi-4. With the new APIs, web developers will be able to add prompt boxes and offer writing assistance tools for text generation, summarizing, and editing. And within the next couple of months, Microsoft says it will also release a text translation API. Microsoft is putting these "experimental" APIs forth as potential web standards, and in addition to being cross-platform, it says they'll also work with other AI models. Developers can start trialing them in the Edge Canary and Dev channels now, the company says. Google offers similar APIs for its Chrome browser. With them, developers can use Chrome's built-in models to offer things like text translation, prompt boxes for text and image generation, and calendar event creation based on webpage content.

Web Apps Can Soon Leverage Microsoft's On-Device AI on Edge

Developers can use the APIs to assist in writing and generating/summarizing text. Artificial Intelligence is now one of the major selling points in various software products, and is slowing making its way into the core functions. While firms continue showcasing new capabilities, web browsers are where AI's arrival is most imminent. Google has already added Gemini Nano to Chrome, and now Microsoft is opening its on-device AI to Edge. Microsoft has announced in the Build conference that it will soon open the existing on-device AI models to web applications, so developers can take advantage of the same. Specifically, the model Phi-4-mini, which was trained on the data from DeepSeek R1's reasoning model. For comparison, it's a close competitor to OpenAI's o3-mini model. These new APIs will help developers develop new apps or add AI capabilities to their existing web applications. They also include a set of APIs to assist in writing and generating/summarizing text. Microsoft also claims that the APIs offer on-device processing and are extremely secure. Microsoft says the APIs will be cross-platform and will also work with other AI models in tandem. They're now available on Edge's Canary and Dev channels. In comparison, Google's Gemini Nano in Chrome also offers a host of features, including text translation and image generation. This is not the only AI-centric announcement from Microsoft this year. The giant also launched Copilot Search in Bing during the Copilot event and its 50th anniversary, which the firm claims will revolutionize search. The Redmond giant is also working to bring a host of AI tools and abilities to Windows with its Copilot+ PCs. What are your thoughts on Microsoft opening its AI models to Edge for web developers. What are some of the apps that you want to see leverage these models? Let us know in the comments.