Google Releases A2UI 0.9 for AI Agent UIs
Google announced the launch of A2UI version 0.9 on April 19, 2026. This framework-agnostic standard supports generative user interfaces. AI agents can use it to create UI elements in real time. They draw from components already present in applications that run on web, mobile, and various other platforms.
Key Components in the Release
The version 0.9 update comes with a shared web core library. It also includes an official renderer for React. Developers see updated renderers for Flutter, Lit, and Angular. Google maintains Flutter as its UI toolkit for cross-platform mobile apps, first released in 2017. Angular serves as a platform for building web applications, led by Google engineers since 2010. React remains a popular JavaScript library for user interfaces, originally developed by Facebook. Lit offers a lightweight way to build web components.
These tools make A2UI compatible with major frontend frameworks. The standard ensures AI agents produce consistent UIs without tying to one specific technology stack.
Agent SDK and New Capabilities
A new Agent SDK simplifies development workflows. It installs via Python right now. Versions for Go and Kotlin will follow soon. Go provides a programming language for scalable software, created by Google in 2009. Kotlin works well for Android development, endorsed by Google since 2017.
The update brings client-defined functions. It adds data syncing between client and server. Error handling sees improvements too. These features help developers build reliable AI-driven interfaces.
Growing Ecosystem and Integrations
Google notes quick growth in the ecosystem around A2UI. Integrations now cover AG2, A2A 1.0, Vercel's json-renderer, and Oracle's Agent Spec. Vercel focuses on frontend deployment and serverless functions, founded in 2015. Oracle delivers enterprise database and cloud services, with a long history in business software dating back to 1977.
Stay ahead of the AI curve
The most important updates, news, and content — delivered weekly.
No spam. Unsubscribe anytime.
Such connections show broad industry interest. They allow A2UI to fit into existing agent protocols and rendering tools.
Early Applications and Resources
Sample applications demonstrate practical uses. Rebel App Studio built a Personal Health Companion. Very Good Ventures created a Life Goal Simulator. Rebel App Studio specializes in custom mobile and web apps. Very Good Ventures offers Flutter expertise and app development services.
Full documentation and code examples sit at A2UI.org. Developers can access them to start building right away.
Google positions A2UI as a key step for AI agents to interact with users through dynamic interfaces. The standard pulls from established components, which reduces custom work. It supports multiple platforms from the start. As AI agents handle more complex tasks, tools like this become essential. The Python-based SDK lowers the entry barrier for many developers. Future Go and Kotlin support will reach backend and mobile specialists.
The integrations with AG2 and A2A 1.0 point to compatibility with other agent standards. Vercel's json-renderer aids web deployments. Oracle's Agent Spec connects to enterprise environments. These ties build a connected network for AI development.
Early apps like the Personal Health Companion show health-related uses. Users might track fitness or medical data via AI-generated screens. The Life Goal Simulator could help with planning and visualization. Such examples prove the standard's flexibility.
Overall, A2UI version 0.9 advances how AI agents present information to people. It relies on proven UI libraries and frameworks.