Overview
LobeHub (the project formerly known as Lobe Chat) is an open-source web app that turns AI models into a chat interface where agents are the unit of work. Instead of one-off prompts in a single window, you create agents, connect them to tools, and have them run on a schedule or collaborate in groups. It is built with TypeScript on a Next.js stack and can be self-hosted.
It is aimed at both end users who want a single place to talk to many models and at developers who want to build and host their own agent setup. You configure an agent once with the Agent Builder, give it access to skills, and reuse it across conversations rather than rebuilding context each time.
As a chat UI in the App UIs and Frontends category, LobeHub focuses on the front end: a polished multi-model chat experience plus a library of MCP-compatible plugins and one-click deployment options, so you can stand up your own hosted instance without writing the interface yourself.
What it does
- Treats agents as reusable units of work: create them with the Agent Builder, schedule them, and review their output
- Multi-model and multi-modal support, so you can route conversations to different models from one interface
- Library of 10,000+ tools and MCP-compatible plugins to give agents access to outside skills
- Agent Groups for running several agents in parallel on a task and iterating together
- One-click self-hosting via Vercel, Zeabur, Sealos, or Alibaba Cloud, plus a Docker Compose setup
- Open-source TypeScript / Next.js codebase you can run and modify yourself
Getting started
The quickest way to try LobeHub is a one-click cloud deploy, or you can self-host with Docker Compose. Both paths need an OpenAI API key set as an environment variable.
Deploy to the cloud (one click)
The README provides deploy buttons for Vercel, Zeabur, Sealos, and Alibaba Cloud. Pick a provider, then set your OpenAI API key in its environment-variable settings (the OPENAI_API_KEY variable is required).
Or self-host with Docker
Create a folder for the database files, run the setup script, then start the stack with Docker Compose. These commands come from the README's Self Hosting section.
mkdir lobehub-db && cd lobehub-db
bash <(curl -fsSL https://lobe.li/setup.sh)
docker compose up -dSet your API key
Provide your model provider key as an environment variable so agents can call the model. OPENAI_API_KEY is required; OPENAI_PROXY_URL and OPENAI_MODEL_LIST are optional.
OPENAI_API_KEY=sk-...Open the app
Once the container is running, open the app in your browser, then use the Agent Builder to describe and create your first agent.
Commands and code are distilled from the project's own documentation — always check the official repo for the latest.
When to use it
- Self-host a private, multi-model chat interface for your team instead of using a single vendor's web app
- Build reusable AI agents with the Agent Builder and connect them to MCP-compatible tools
- Run several agents in an Agent Group to work on a task in parallel and refine the result
- Compare or switch between different models from one chat UI without rebuilding the front end
How Lobe Chat compares
Lobe Chat alongside other open-source chat uis tools AI/TLDR tracks, ranked by GitHub stars.
| Tool | Stars | What it does |
|---|---|---|
| Open WebUI | ★ 142k | A self-hosted web interface for chatting with local and API-based LLMs (Ollama, OpenAI-compatible) that adds RAG, web search, image generation, and multi-user access. |
| Lobe Chat | ★ 78.9k | Open-source chat UI for building and running a team of AI agents |
| AnythingLLM | ★ 61.8k | A desktop and self-hosted chat app focused on document Q&A and agents, letting you build private RAG workspaces over your own files and data sources. |
| Cherry Studio | ★ 47.5k | A cross-platform desktop chat client that connects to many cloud and local model providers and ships with hundreds of preset assistants and MCP support. |
| Chatbox | ★ 40.6k | A desktop and web chat client for ChatGPT, Claude, Gemini, Ollama, and other models, with local message storage and multi-provider switching. |
| LibreChat | ★ 39.5k | A self-hostable ChatGPT-style chat app that unifies many AI providers and adds agents, MCP support, artifacts, a code interpreter, and multi-user authentication. |
| Onyx | ★ 30.4k | Onyx is a self-hostable AI chat platform that adds RAG, web search, custom agents, code execution, and deep research on top of any major LLM provider. |
| Vercel AI Chatbot | ★ 20.5k | An open-source Next.js starter template for building a full-featured AI chatbot with the Vercel AI SDK, auth, and multi-model routing. |