wowsuch ai
Live

GoogleDeveloper platform

A2UI

A2UI is a framework-agnostic open standard for describing generative user interfaces as data for an application to render with its existing components. Its declarative format uses trusted component catalogs and supports incremental streaming of JSON messages.

What it does

Developers can use A2UI to let agents compose interactive interfaces from their applications’ existing components.

Availability
Available in the Python agent SDK, TypeScript agent SDK, and Dart SDK
Pricing model
Not established
Deployment
The client renders A2UI operations using its component catalog
Privacy
Not established
Licence
Described as an open standard
Capability
agentsautomationtext generation
Modality
textmultimodalcode
Access
SDK
Deployment
on-deviceon-premises
Pricing
unknown
Target user
developer

Evidence

11 claims
  • A2UI describes interfaces as data for applications to render with their existing components. [1]
  • A2UI is a declarative data format rather than executable code. [1]
    A2UI is a declarative data format rather than executable code.
    Google. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • The client keeps a catalog of trusted components and checks agent output against it. [1]
    The client keeps a catalog of trusted components, so anything the agent invents outside that set is caught before it reaches the screen.
    Google. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • A2UI payloads are flat lists of JSON messages that can be emitted incrementally. [1]
  • The same response can render through React, Lit, Angular, SwiftUI, Jetpack Compose, or Flutter. [1]
    The same response renders through React, Lit, Angular, SwiftUI, Jetpack Compose (Android) or Flutter (multi-platform), and through community renderers for Vue, or even the terminal.
    Google. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • The article provides setup instructions and links to A2UI documentation.
  • A2UI Express is an inference format optimized for inference speed and accuracy. [2]
    Now agents can generate A2UI Express , an intermediate format optimized for inference speed and accuracy, and rely on deterministic code in A2UI's Agent SDKs to translate it into the JSON-based transport protocol.
    Google. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • The SDK compiles the model's reply into validated A2UI messages. [2]
    The SDK generates the prompt instructions for that format and compiles the model's reply into standard A2UI messages.
    Google. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • The inference format and compiler are generated from the catalog's JSON schema. [2]
    An inference format has two parts, and the SDK generates both from your catalog's JSON schema.
    Google. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • A2UI Express is available in the Python, TypeScript, and Dart SDKs. [2]
    A2UI Express is available in the Python agent SDK ( a2ui-agent-sdk 0.9.0 on PyPI), in the TypeScript agent SDK ( @a2ui/agent on npm), and in the Dart SDK ( a2ui_agent 0.0.1-wip004 on pub.dev).
    Google. Quoted text is outside CC BY 4.0 and keeps its original rights.
  • The post reports that Express can reduce latency by up to 3x and use up to 3x fewer tokens than Direct JSON for the same kinds of UI. [2]

Sources

2
1
Verified
2
GoogleOwner-controlled
https://developers.googleblog.com/fast-reliable-a2ui/ Retrieved 2026-10-28 00:19 UTC
Verified

Versions

2
VersionEffectiveChangeChanged fieldsState
22026-10-30 00:39 UTCmajor releasemodality, capability, deployment, availability, releaseCurrent
12026-10-30 00:38 UTCpublicationNoneCurrent

Open in the catalog

Original catalog content is licensed under CC BY 4.0, attributed to wowsuchai. Source quotations and linked content keep their original rights.