Lobe Chat: Open-Source Local AI Chat Framework with Multi-Model

Lobe Chat is an open-source AI chat framework built on Next.js and TypeScript, enabling one-click private deployment of ChatGPT, Claude, Gemini, Ollama, and more through multiple AI providers.
Featuring knowledge base integration, RAG, file uploads, and a function-calling plugin system, it empowers developers and end-users with a secure, extensible, and modern Material You inspired interface.

How to Use

1. Clone or Preview:

   • Clone via Git: git clone https://github.com/lobehub/lobe-chat.git

   • One-click demo: Visit chat-preview.lobehub.comGitHub

2. Install & Run:

   cd lobe-chat
   pnpm install
   pnpm dev

3. Configure Providers:
   Copy .env.example to .env, fill in OPENAI_API_KEY, CLAUDE_API_KEY, etc., then restart the server.

4. Production Deployment:

   • Vercel/Zeabur/Sealos: Connect your repo, set environment variables, and enable auto-deploy.

   • Docker: Run docker-compose up -d for a turnkey private chat service.

5. Access UI:
   Navigate to http://localhost:3000 to start your secure AI chat application.

Features

• Multi-Provider Support: Seamlessly switch between OpenAI, Claude 3, Gemini, Ollama, DeepSeek, and Qwen APIs.

• Knowledge Base & RAG: Upload documents, manage knowledge, and leverage Retrieval-Augmented Generation for context-aware answers.

• Plugin & Chain-of-Thought: Extend functionality via a robust plugin system supporting function calls, branching dialogues, and artifacts.

• Local LLMs: Run local models like Llama 2 and Ollama offline to enhance privacy and reduce latency.

• TTS & STT: Built-in text-to-speech and speech-to-text modules enable natural voice interactions.

• PWA & Mobile Adaptation: Progressive Web App support ensures smooth usage on smartphones and tablets without installation.

• Custom Themes: Light/dark mode switch and brand-color theming options for personalized UIs.

Use Cases

• Enterprise Knowledge Q&A: Build a private, secure chatbot for internal documentation and customer support via RAG and file uploads.

• Dev Prototype & Testing: Quickly evaluate multiple LLM providers in one unified framework for faster iteration cycles.

• Education & E-Learning: On-premise deployment ensures data privacy for students and instructors while enabling interactive lessons.

• Personal Writing Assistant: Leverage chain-of-thought reasoning and multi-modal outputs to boost creativity and writing efficiency.

• Cross-Platform Chat: Use PWA and voice chat to provide seamless AI communication experiences across devices.