Google has officially released version 1.0 of A2UI, an open-source protocol designed to enable AI agents to generate user interface elements directly. The protocol allows agents to describe content, which is then rendered using an application's native components to ensure high fidelity and low latency.
The 1.0 release introduces bidirectional function calling and expanded support for accessibility and metadata. Developers can utilize SDKs for Python, TypeScript, and Dart to build agents, alongside open-source renderers for web, Android, and iOS platforms.
The rendering ecosystem supports various frameworks, including React, Angular, and Lit for web, as well as SwiftUI for iOS and Jetpack Compose for Android. Flutter support is also available for mobile, web, and desktop environments.
The protocol is compatible with Model Context Protocol (MCP) applications, allowing A2UI content to be displayed within AI assistants such as Gemini and Claude. Major companies, including ByteDance, Bloomberg, Atlassian, and Alation, have already adopted the protocol for development.
