How OpenUI launched on X

OpenUI is open-source infrastructure that lets AI apps and agents respond with interactive UI, charts, tables, forms, and dashboards instead of plain text . It targets developers who want to move their LLM products past the chat textbox without rebuilding their stack, and it is designed to work with any model (OpenAI, Anthropic, Gemini, DeepSeek, xAI, Mistral, Qwen, Ollama), any agent framework (Vercel AI SDK, LangChain, LangGraph, Convex, Cloudflare, CrewAI, n8n, Haystack), any design system (shadcn/ui, Material UI, Ant Design, Radix UI, Chakra UI, Mantine, daisyUI), and any web framework such as React, Next.js, Vue, Svelte, Angular, Nuxt, Remix, or Astro . At the center of the launch is OpenUI Lang, a compact, streaming-first language that LLMs write to build interfaces from your own components, where each line defines one component so the interface renders while the response is still streaming . OpenUI Lang generates up to 3x faster and uses 67% fewer tokens than JSON-based UI formats , and the team pairs it with OpenUI Gateway, which catches and repairs invalid output before users see it and switches providers during outages to keep the same model online . The tweet notes the project was open-sourced six months ago and has crossed 10K GitHub stars, which aligns with current repo traction. OpenUI is built by Thesys, where Thesys is the company behind OpenUI, much like Vercel is the company behind Next.js, maintaining the open-source framework and operating OpenUI Cloud . The company is led by co-founder and CEO Rabi Shanker Guha, previously Head of Engineering (India) at DevRev and an engineer at Google , alongside co-founder Parikshit Deshmukh. For teams betting on agents, the launch matters because it offers a production path for generative interfaces that plugs into existing components, models, and infra rather than forcing a rewrite.