Saturday, 5 July 2025
27.9 C
Singapore
30 C
Thailand
20.3 C
Indonesia
28.8 C
Philippines

Microsoft brings on-device AI to web apps in the Edge browser

Microsoft adds on-device AI to Edge, giving web apps access to Phi-4-mini for smart features like text editing and translation.

If you’re a web developer, you’re about to get access to a powerful new tool. Microsoft has announced that you’ll soon be able to use on-device AI models directly in the Edge browser. Revealed at the company’s Build conference on May 20, this update means your web apps can connect with Microsoft’s AI model, the Phi-4-mini, using new programming interfaces, or APIs.

This move isn’t just for Windows users, either. Microsoft confirmed that these APIs will work across different platforms, including macOS. So, no matter which system you’re using Edge on, you’ll be able to benefit from this new AI functionality.

What is Phi-4-mini?

The Phi-4-mini is Microsoft’s latest lightweight AI model. It’s built with 3.8 billion parameters and was launched in February alongside its larger counterpart, the Phi-4. Even though it’s small compared to some of the bigger models, Phi-4-mini is designed to run directly on your device. This means faster responses, better privacy, and no need for internet access to run basic AI tasks.

With the help of the new APIs, you can now use this model in your web projects to create features such as text boxes with smart prompts, writing assistance tools, and options for summarising or editing text. These tools are helpful for writing apps, chatbots, or any other web-based tool that benefits from smart content generation.

And that’s not all—Microsoft also mentioned that a new text translation API is in the works and should be available in the coming months.

AI tools that go beyond Microsoft

Microsoft is offering these APIs as experimental tools. The goal is to eventually make them part of a broader set of web standards. The company has said that the tools will work not only with Phi-4-mini but also with other AI models. This opens the door for more flexible development and allows you to choose the model that best fits your needs.

If you’re interested in testing these tools immediately, you can find them in the Edge Canary and Dev versions. These early-release browser versions let you try out new features before officially launching.

Microsoft isn’t alone in this space. Google already provides similar AI features in its Chrome browser. Chrome’s APIs allow web apps to use Google’s built-in models for tasks like translating text, generating content, and even creating calendar events based on information found on a web page.

What this means for developers

This change is part of a growing trend where AI tools are becoming more available at the browser level. For you as a developer, it means you don’t always need cloud access to bring smart features into your apps. On-device models offer a faster, more secure way to use AI.

By letting your apps connect directly with tools like Phi-4-mini, you’ll be able to give users real-time writing support, faster translations, and a smoother overall experience. This big step forward could shortly reshape the way web apps are built.

Hot this week

Google launches Gemini AI for schools and students, raising questions about future of learning

Google launches Gemini AI in schools with safety tools and fact-checking, sparking debate on its impact on learning and student development.

Mainland investment boom lifts Hong Kong’s market

Chinese firms turn to Hong Kong listings after mainland investors spend US$93B on stocks, eyeing global growth and fresh funding sources.

Google lets you share smart home access more easily with family and kids

Google Home lets you easily assign Admin or Member roles, even for kids under 13, to manage your smart home access better.

Kahoot! sets up Singapore office to support APAC expansion

Kahoot! opens Singapore office as APAC hub to support growing demand for digital learning in education and corporate sectors.

Self-driving shuttles to begin rolling through Punggol by late 2025

Self-driving shuttles will launch in Punggol by late 2025, bringing autonomous public transport to Singapore neighbourhoods for the first time.

DeepSWE, powered by Alibaba’s Qwen3-32B, outperforms rivals in global benchmark

Alibaba’s open-source Qwen model powers DeepSWE to global victory in AI agent rankings, signalling a shift in open-weight AI innovation.

E Ink transforms laptop touchpads into smart e-reader displays for AI use

E Ink’s new touchpad brings e-reader tech to laptops, offering a low-power screen for AI apps and assistants right under your fingertips.

Xiaomi opens new store at City Square Mall and launches Shopee presence in Singapore

Xiaomi opens its ninth store in Singapore at City Square Mall and launches its official Shopee store with promotional offers across both platforms.

Tools for Humanity: Why Southeast Asia is shaping the future of humanness in the Age of AI

Southeast Asia is pioneering the future of digital identity with World ID, offering private, secure, and human-first verification at scale.

Related Articles

Popular Categories