What is generative UI?

Generative UI is interfaces produced by language models at runtime. Components are streamed from LLM output and rendered as real UI, not text. The model picks the form alongside the content: a chart for numbers, a form for input, a list for choices.

Traditional UI
  • Form chosen by the designer at build time
  • Static templates, deterministic outcomes
  • Same screen for every user, every request
  • Changes ship through a release cycle
Generative UI
  • Form chosen by the model at inference time
  • Runtime components, probabilistic outcomes
  • Different shape per user, per request
  • Changes ship by editing the prompt or the catalog

State of generative UI — runtimes at a glance

Five production-grade options, compared on stack, streaming model, openness, and maturity.

Last updated 2026-03-08

Vercel AI SDK

Mature

TypeScript · RSC / Vue

Streaming React/Vue components with first-class tool use. The default for new Next.js and Nuxt apps.

CopilotKit

Maturing

TypeScript · React Hooks

Drop-in copilot hooks and UI primitives. Strong for adding an assistant to an existing app.

Tambo

Early

TypeScript · React catalog

Component catalog the model picks from. Useful when the design system is the unit of generation.

Thesys C1

Maturing

Hosted API · any client

Hosted API that returns ready-to-render UI. Trades infra ownership for time-to-first-screen.

A2UI (Google)

Early

JSON spec · renderer-agnostic

Open JSON schema for renderer-agnostic UI from agents. Cross-runtime, very early.

Código abierto

Markdown en streaming sin parpadeos

Los renderizadores basados en innerHTML destruyen el DOM en cada token, provocando parpadeos, pérdida de foco y saltos de layout. Generative DOM compara el AST y actualiza solo lo que cambió: hasta 90× menos trabajo en el DOM, sin parpadeos.