A2UI 1.0

A2UI 1.0 is the first stable version of Google’s open protocol for generative UI, announced on November 11, 2026, in the Google Developers Blog post A2UI 1.0: Build delightful, fast, reliable, agentic experiences, by Vishal Motwani, A2UI PM, and the Google A2UI Team. The post says: “Today, we're releasing A2UI 1.0, the first stable version of our open protocol for generative UI.” It lets AI agents answer with interactive screens, such as a form to confirm or a chart, instead of a wall of text. Developers can use it now with open-source renderer SDKs for the web, Android, and iOS, and agent SDKs for Python, TypeScript, and Dart. The post says companies such as Atlassian, Alation, Bloomberg, and ByteDance are already building with it. The site is a2ui.org, and the code is at github.com/a2ui-project/a2ui.

Officially, the agent describes what to show, and every app draws the UI with its own native components. The application receives A2UI messages and renders them securely at native speed. The post says users get the same low latency, high trust, and native fidelity whether they reach a company’s agent in its own app or in a third-party AI assistant that hosts MCP apps, such as Gemini or Claude.

What is new in 1.0, per the official post, is in three parts. A stable protocol: version 1.0 adds bidirectional function calls, so a flight or reservation picker can check availability on the spot, plus support for accessibility and extensions for additional metadata. Renderer libraries for major platforms: open-source renderers draw A2UI screens with React, Angular, and Lit on the web; Flutter on mobile, web, and desktop; SwiftUI on iOS; and Jetpack Compose on Android (currently v0.9.1, with v1.0 coming soon). The web and Flutter renderers also work inside MCP Apps, so A2UI screens can appear in AI assistants that support MCP Apps. Agent SDKs: new SDKs for TypeScript and Dart, along with Python, simplify integration by generating skills, translating messages, and correcting errors. Swappable inference formats reduce latency, and in Python, A2UI Macros add programmatic component expansion and data binding.

On adopters, the official post says Atlassian is bringing A2UI to Jira, where an agent can ask the team for a missing detail or a decision with an interactive card on the work item; Alation’s Alation Intelligence Operating System (AIOS) builds feeds of metric cards and charts with A2UI, and the same feed can be viewed in Claude and inside Alation with a nearly identical layout; ByteDance released an A2UI renderer for its cross-platform framework Lynx in June 2026, and Lynx supports A2UI 0.9 today while adapting to 1.0 with the A2UI team. Inside Google, Gemini Notebook uses A2UI cards to ask about slide-deck format and topics and to let you edit a review card before generation. Organizations can register their own agents that respond with A2UI; Gemini supports A2UI 0.9.1 today. The post also says agents in Gemini Enterprise can already use A2UI, and the team is working on upgrading Gemini Enterprise to A2UI 1.0. The official Google Cloud Tech account @GoogleCloudTech posted at 2026-11-11T17:03:51Z: “Today we are releasing A2UI 1.0, now faster, easier, and stable, an open protocol and framework for generative UI across web, Android, iOS, wearables, TVs, and any other surface.”

Key points:

• Official blog: https://developers.googleblog.com/en/a2ui-10-build-delightful-fast-reliable-agentic-experiences/ ; site: https://a2ui.org ; code: https://github.com/a2ui-project/a2ui ;
• Announcement post: https://x.com/GoogleCloudTech/status/2120563165557571629 (2026-11-11T17:03:51Z);
• This is a specific Google protocol version name, the first stable release of A2UI. Do not describe it as still being the 0.9 / 0.9.1 preview, or confuse it with MCP Apps itself.

Why it matters

This is a formal stable-version name Google published on November 11, 2026, in an official developers blog post and from the Google Cloud Tech account, not a rumor. A2UI already existed as 0.9 / 0.9.1, but 1.0 is what the official post calls the first stable version, with bidirectional function calls, multi-platform renderers, and new agent SDKs. For anyone tracking generative UI and agent interface protocols, this is a new name that deserves its own entry.

This page only restates public official materials. Partner statements, the Lynx support status, and phrases such as “faster, easier, and stable” come from Google or from partners Google quotes; this page does not independently evaluate them.

Limits

Public materials say the Jetpack Compose renderer for 1.0 is still forthcoming (it supports 0.9.1 today), Gemini supports A2UI 0.9.1 today, and the upgrade of Gemini Enterprise to A2UI 1.0 is still in progress. A stable protocol is not the same thing as every host app and renderer already being on 1.0. Do not treat this entry as a claim that every assistant already natively supports A2UI 1.0.

1. Read the official blog at https://developers.googleblog.com/en/a2ui-10-build-delightful-fast-reliable-agentic-experiences/ , and compare it with the site at https://a2ui.org , the code at https://github.com/a2ui-project/a2ui , and the Google Cloud Tech announcement at https://x.com/GoogleCloudTech/status/2120563165557571629 ;
2. Follow the official quickstart to pick a renderer (React, Angular, and Lit on the web; Flutter; SwiftUI on iOS; Jetpack Compose on Android is 0.9.1 today) and an agent SDK (Python, TypeScript, Dart); if you are upgrading from 0.9, use the official evolution guide;
3. When you cite “first stable version,” bidirectional function calls, partner examples, or platform support, attribute them to the 2026-11-11 official blog or announcement post and re-check the upstream source;
4. Do not describe A2UI 1.0 as still being the 0.9 preview, or confuse it with the MCP Apps protocol itself; this page only restates the official blog and announcement, so re-check the upstream source before you cite it elsewhere.