Analysis updated 2026-06-20
Build a private ChatGPT-style chat app for your team using your own API keys
Self-host a multi-model AI chat interface so your conversation data stays on your own servers
Connect to locally running AI models via Ollama for fully offline AI chat
Deploy a branded AI assistant with your own database and authentication
| mckaywrigley/chatbot-ui | backstage/backstage | immutable-js/immutable-js | |
|---|---|---|---|
| Stars | 33,206 | 33,285 | 33,061 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | moderate | hard | easy |
| Complexity | 3/5 | 4/5 | 2/5 |
| Audience | vibe coder | ops devops | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires a Supabase project (or local Docker setup) plus at least one AI provider API key before the app will work.
Chatbot UI is an open-source web interface for chatting with AI language models. It solves the problem of vendor lock-in and privacy: instead of being tied to a single provider's chat interface, you can run this app yourself and connect it to any AI model you choose, including OpenAI's GPT models, Azure OpenAI, or locally running models via Ollama (a tool for running AI models on your own computer). The interface looks and works much like popular AI chat apps: you have conversations, can manage multiple chat threads, and can switch between models. Because you self-host it, your conversations stay in your own database rather than a third-party service's servers. The way it works is that you deploy the frontend (a Next.js web app) and connect it to Supabase, which is an open-source Postgres database service, for storing conversations and user accounts. You provide your own API keys for whichever AI models you want to use. The application routes your messages to the appropriate AI provider and displays the responses in the chat interface. You would use Chatbot UI when you want a polished, multi-model chat interface but want control over your data, want to use locally running models, or want to avoid paying for a hosted AI app subscription when you already have API access. It is also useful for building a private ChatGPT-style app for a team. The tech stack is TypeScript and React on the frontend using Next.js, Supabase (Postgres) as the database and authentication backend, and it can be deployed to Vercel for hosting. Local development requires Docker to run Supabase locally. An official hosted version is available at chatbotui.com for users who don't want to self-host.
A self-hosted, open-source chat interface that lets you talk to any AI model, OpenAI, Azure, or locally running models, while keeping your conversations in your own database.
Mainly TypeScript. The stack also includes TypeScript, React, Next.js.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.