Assistant UI
assistant-ui/assistant-ui
Componentes React prontos para chat com IA em produção
Ver no GitHubBiblioteca TypeScript/React open-source para construir experiências de chat com IA em produção. Oferece componentes compostos (Thread, Message, Composer, etc.) prontos para estilização customizada ou com tema pré-configurado do shadcn/ui. Integra nativamente com Vercel AI SDK e suporta runtimes customizados.
Instalação
npx convez-ai assistant-ui/assistant-ui1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Construir interfaces de chat personalizadas em aplicações React/Next.js
- Integrar IA conversacional com streaming e suporte a ferramentas/actions
- Prototipação rápida com CLI scaffolding ou adição incremental a projetos existentes
Veredito

Solução madura e bem-arquitetada para quem precisa de UI de chat em React sem reinventar a roda. A composição de primitivos e suporte a múltiplos runtimes (Vercel AI SDK, LangGraph, custom) oferece flexibilidade real, não marketing. Backed by Y Combinator e com comunidade ativa.
README
A tradução para o português deste README ainda está na fila.
The UX of ChatGPT in your React app 💬🚀
assistant-ui is an open-source TypeScript/React library to build production-grade AI chat experiences fast.
Installation
The fastest path is the CLI, which scaffolds a Next.js app or adds the styled components to an existing project:
npx assistant-ui@latest create # new project
npx assistant-ui@latest init # add to existing project
Or install the packages directly:
npm install @assistant-ui/react @assistant-ui/react-ai-sdk
Usage
"use client";
import { AssistantRuntimeProvider } from "@assistant-ui/react";
import { useChatRuntime } from "@assistant-ui/react-ai-sdk";
import { Thread } from "@/components/assistant-ui/thread";
export function Chat() {
const runtime = useChatRuntime();
return (
<AssistantRuntimeProvider runtime={runtime}>
<Thread />
</AssistantRuntimeProvider>
);
}
useChatRuntime connects to the Vercel AI SDK out of the box. Swap it for useLangGraphRuntime, useDataStreamRuntime, or any custom runtime to integrate with your own backend.
What you get
- Composable primitives: build any chat UX from
Thread,Message,Composer,ThreadList,ActionBar, and friends. Style every pixel yourself, or start from a polished shadcn/ui theme that the CLI copies into your project. - Production UX out of the box: streaming, auto-scroll, retries, attachments, markdown, code highlighting, voice dictation, keyboard shortcuts, and accessibility.
- Generative UI: render tool calls and JSON as React components, collect inline human approvals, and expose safe frontend actions to the model.
- Strong TypeScript: typed runtime APIs, tool schemas, message parts, and adapters end to end.
Backends
| Integration | Package |
|---|---|
| Vercel AI SDK | @assistant-ui/react-ai-sdk |
| LangGraph / LangChain | @assistant-ui/react-langgraph, @assistant-ui/react-langchain |
| AG-UI / A2A protocols | @assistant-ui/react-ag-ui, @assistant-ui/react-a2a |
| Google ADK / OpenCode | @assistant-ui/react-google-adk, @assistant-ui/react-opencode |
| Custom data-stream backend | @assistant-ui/react-data-stream |
| Managed thread history, telemetry, and file storage | assistant-cloud |
Broad model support out of the box (OpenAI, Anthropic, Google Gemini, Mistral, Perplexity, AWS Bedrock, Azure, Fireworks, Ollama) plus community providers via the AI SDK, and easy extension to any custom HTTP backend.
Customization
Radix-style: instead of a single monolithic chat component, you compose primitives and bring your own styles. The CLI ships a great starter; you control everything else.

Sample customization to make a Perplexity lookalike:

Used in production by
, , , , , , , , , , and many more.

Demos
Community & Support
For other platforms
- React Native:
@assistant-ui/react-native - Terminal (Ink):
@assistant-ui/react-ink
License
MIT, with optional Assistant Cloud for managed thread persistence and analytics.
Backed by Y Combinator.
