Artificial Intelligence1 min read

Hybrid intelligence on the Web: Introducing Aion Instruct and Prompt API in Microsoft Edge

By · Published by Everything Blog

In short

Microsoft has launched the 2.3-billion-parameter Aion Instruct model to serve as the default on-device engine for Edge, replacing the larger Phi-4-mini. The integration is accompanied by an expanded origin trial for the Prompt API, giving developers the tools to build and test local generative tasks directly on live websites. An experimental Decisions API is also entering preview to compute single-pass structured evaluations without text generation overhead. Furthermore, the browser only downloads the model when required by sites on capable hardware, and users retain complete control to toggle the feature off.

Key points

  • Web creators can adopt a hybrid model architecture that prioritizes fast local execution…: Web creators can adopt a hybrid model architecture that prioritizes fast local execution while reserving cloud resources for heavier computing needs.
  • Microsoft plans to extend the Aion Instruct framework into Windows to establish a baselin…: Microsoft plans to extend the Aion Instruct framework into Windows to establish a baseline AI layer across Copilot+ PCs later this year.
  • Edge protects user control and system storage by only downloading the underlying model on…: Edge protects user control and system storage by only downloading the underlying model on capable hardware when specific websites require it.

Sources checked by our analysis pipeline: 1

Rapidly evolving AI capabilities are redefining what sites can do for users on the web. With on-device AI in Microsoft Edge, sites can use task-specific and small language models for capabilities such as writing assistance, summarization, language detection, translation, and speech recognition. Through the Prompt API, they can also use a small language model for more general-purpose intelligence.

Together, these capabilities provide a foundation for hybrid intelligence on the web: developers can leverage on-device AI models when they are available in the browser and well-suited, and switch to a configured cloud provider to extend device coverage or access more powerful inference when needed.

To make on-device inference more broadly available, we’re introducing three updates today:

- General availability of Aion Instruct as an open-weights model on Hugging Face, and as Microsoft Edge’s built-in small language model.

- An origin trial for the Prompt API, so you can use the Prompt API and Aion Instruct on your live site today.

- A developer preview of an experimental Decisions API, which gets you fast, structured answers about text.

These announcements build on…

Artificial Intelligence