OpenUI
Open standard and toolkit for AI models to generate and stream live user interfaces.
- Stars
- 14.7k
- Forks
- 961
- Licence
- MIT
- Language
- TypeScript
- Last update
- 4 days ago
- Type
- Framework
What it is
OpenUI is a framework for building interfaces that an AI model generates on the fly. It defines a compact language, OpenUI Lang, that a model streams as it writes, and a renderer that turns that stream into live UI in React, Vue, Svelte, or Angular. You define a set of components, and OpenUI builds the instructions that tell the model which components it may use.
- Scaffold a starter chat app with npx @openuidev/cli@latest create --name genui-chat-app
- Define a component library and generate a system prompt from it for your model
- Stream model output and render charts, tables, or forms progressively as tokens arrive
- Embed a prebuilt renderer via @openuidev/browser-bundle in a page without a build step
Great for
- Developers adding AI-generated charts, forms, or tables to a web app
- Teams building chat assistants or copilots that answer with interactive UI
- React developers who want prebuilt chat layouts and component libraries
- Developers using LangChain or LangGraph agents that stream through AG-UI
Not ideal for
- People who only need a plain chatbot with text replies, since the project is built around generated UI
- Teams not using a supported frontend (React, Vue, Svelte, Angular) without extra integration work
- Anyone who needs a fully proven production track record, since the README does not describe production deployments
How to get it
- 1
Install with a package manager
You'll need: Node.js (version not stated in the README)
Scaffold a new chat app
npx @openuidev/cli@latest create --name genui-chat-app - 2
Run it
- Run the scaffold command above to create a project folder
- Open the generated project and follow its own README to start the dev server
- Try generating UI live in the Playground at openui.com/playground
- Define or reuse a component library and generate a system prompt from it
- 3
Get the most out of it
- Restrict the model to a small, well-defined component library to get more predictable output
- Use @openuidev/react-headless if you want your own chat UI on top of the state logic
- Use @openuidev/langchain to stream OpenUI through AG-UI from LangChain or LangGraph agents
- Use @openuidev/cli to generate a system prompt or JSON schema from your library definition
Downloads
Strengths
- MIT licensed and open source, with a published package set for several frontends
- Claims up to 67% fewer tokens than JSON for the same output (a figure from the README, not independently verified here)
- Streams output progressively, so the UI updates as the model writes
Trade-offs
- The README is a broad overview; detailed API docs live on the external docs site
- Many packages and integrations make the right starting point unclear for newcomers
- Output quality depends on the model and on how well the component library is defined
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| openui (this) | 14.7k | Framework | intermediate | 4 days ago | Developers adding AI-generated charts, forms, or tables to a web app |
| CopilotKit | 38.5k | Library | intermediate | 4 days ago | Developers adding an AI assistant or agent interface to a React or Next.js app |
| assistant-ui | 12.9k | Library | intermediate | 4 days ago | React or Next.js developers adding an AI chat panel to an existing app |
| tambo | 11.2k | Library | intermediate | 5 days ago | React developers who want an AI chat to show charts, forms, or boards instead of only text |
| langchain | 149k | Framework | intermediate | 4 days ago | Python developers building LLM-powered applications |
| ai-pdf-chatbot-langchain | 16.6k | App | intermediate | 7 months ago | Developers learning how LangGraph ingestion and retrieval graphs fit together |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.