LibreChat

LibreChat is a self-hosted AI chat platform that unifies OpenAI, Anthropic, Google, AWS Bedrock, Azure, and other model providers behind one interface. Its UI takes cues from ChatGPT, then adds agents, Model Context Protocol (MCP) tool support, a sandboxed code interpreter, and multi-user authentication on top. It functions as an open-source alternative to ChatGPT for teams that want to run their own chat platform against multiple model providers.

Key features

• Multi-provider model selection: Switch between Anthropic Claude, OpenAI, Azure OpenAI, AWS Bedrock, Google Vertex AI, and OpenAI-compatible custom endpoints like Ollama, Groq, Mistral, or OpenRouter.
• Agents and MCP tools: Build no-code custom agents, share them through an agent marketplace, and connect Model Context Protocol servers for tool use.
• Code interpreter: Run sandboxed code in Python, Node.js, Go, C/C++, Java, PHP, Rust, or Fortran, with file upload and download built in.
• Web search: Combine search providers, content scrapers, and result rerankers to pull live information into a conversation.
• Generative UI artifacts: Generate React components, HTML, and Mermaid diagrams directly inside the chat.
• Image generation and editing: Create and edit images with GPT-Image-1, DALL-E, Stable Diffusion, Flux, or any connected MCP image server.
• Multi-user administration: Secure multi-user access with OAuth2, LDAP, or email login, plus a browser-based admin panel for managing users, groups, and roles without redeploying.

Self-hosting and deployment

LibreChat is designed to be run entirely on your own infrastructure or in the cloud. One-click deploy templates exist for Railway, Zeabur, and Sealos, and the project ships Docker Compose stacks that bundle the admin panel by default. Media can be served through S3 with CloudFront for stable links, edge delivery, signed cookies, and secured downloads.

Ideal use cases

LibreChat suits organizations that want a single chat interface across several model providers instead of separate logins per vendor, plus teams building internal AI agents that need MCP tool access, code execution, or multi-user access control.

Conversation and interface features

Beyond model access, LibreChat covers the details of daily chat use: conversation branching so you can edit and resubmit earlier messages, message and conversation forking for advanced context control, importing conversations from LibreChat, ChatGPT, or Chatbot UI, and exporting them as screenshots, Markdown, plain text, or JSON. Resumable streams mean an AI response automatically reconnects if your connection drops, and stays in sync across multiple open tabs or devices, scaling from a single server to a horizontally scaled deployment with Redis. The interface itself ships in over 30 languages, from English and Mandarin to Turkish, Hebrew, and Uyghur, and supports speech-to-text and text-to-speech through OpenAI, Azure OpenAI, or ElevenLabs for hands-free use.