The next generation of apps may not have a fixed interface, as Jev enables real-time reorganization of app interfaces.

Beating AI News Flash: Vercel Labs engineer Chris Tate has integrated Jev into json-render, beginning experiments to let AI directly compose UI at runtime.

json-render pre-prepares buttons, cards, input fields, data, and actions. Jev doesn't need to write complete JSON token by token like a large model—it only decides which components to use, how to order them, and where to place them, then hands off to a normal UI renderer. The new interface defaults to selecting primary components in a single pass, and when needed, uses a second pass to arrange the layout.

In the same train ticket interface generation demo, the default JSONL mode took 3.21 seconds from request to completion, while Jev took only 0.88 seconds.

Future app interfaces may not need to be fully hardcoded at all. UI can become a runtime decision, where the same app temporarily assembles different pages for different users, tasks, and states. Jev's demo has already pushed full generation time under 1 second, at least giving this kind of interaction a sense of real-time responsiveness.

Runway's Solaris, released at the end of last month, is also betting on a similar direction, just with a more aggressive approach. Solaris uses a world model to generate the entire interface frame by frame, bypassing traditional code and components entirely; Jev still uses components and actions pre-written by developers, only dynamically deciding how to combine them. One wants to turn software into a real-time generated video world, while the other is more like giving software a face that can be reassembled at any time.