An open source toolkit to make your
AI agents respond with your UI.
- 1
You define
your libraryRegister components with
defineComponent&createLibrary. - 2
OpenUI generates
system promptGenerate a system prompt from your library and send it to LLM.
- 3
LLM responds
in OpenUI LangThe model returns token-efficient, line-oriented OpenUI Lang.
- 4
Renderer parses
and renders UIThe renderer parses the output and renders UI in real time.
Generative UI for every use case
From dashboards to support to dev tools, OpenUI renders the interface your users need.
Analytics
Help users move from raw data to dashboards, insights, and decision-ready reports.
CRM / Sales
Customer support
DevTools
Project management
Help users move from raw data to dashboards, insights, and decision-ready reports.
Live data
Interfaces query your tools and MCP servers at runtime, so the data is always current.
Cross-platform
One spec renders natively to React, React Native, Vue, and more.
Streaming-first
UI renders progressively as the model responds, not after the full output.
Interactive
Reactive state, inputs, and actions wired straight to your tools.
Safe by default
The model only composes your components and never runs arbitrary code.
Bring your own components
Build from the components and design system you already have.
OpenUI works with any stack
OpenUI works with any LLM, UI library, and agent framework.
Add generative UI without changing your stack.
Production-grade rendering
Tested, responsive components that render consistently across browsers, devices, and screen sizes.
Editable artifact generation
Generate static artifacts like slides and reports, and live artifacts like dashboards and pages.
Bring your design system
Apply your fonts, colors, spacing, and component styles across every generated interface.
Output validation
Detect and correct invalid model output before it turns into broken UI.
Model & provider resilience
Keep generated UI working across model quirks, upgrades, slowdowns, and provider failures.
Observability & audit trail
Track performance, failures, cost, and what was rendered across your deployment.
Want to learn more, or need hands-on engineering support?
Talk to our team