Google explains how to speed up A2UI with swappable inference formats
AIGoogle introduces A2UI Express, an inference format that the agent SDKs compile into standard A2UI messages, and says it reduces latency by up to 3x and uses up to 3x fewer tokens than Direct JSON. The post also describes controlled generation, where a decision model called Jev selects among predefined layouts, reducing latency by up to 80x for layout decisions. Express is available in the Python, TypeScript, and Dart agent SDKs.
AIWhy it matters.
The post explains how swapping inference formats cuts A2UI latency and token use, with a concrete SDK setup and tradeoffs between Express and controlled generation.
Source details
Google Developers Blog · developers.googleblog.comPublished · added here
Original title: Fast, reliable A2UI, with swappable inference formats