LibreChat — Self-Hosted Multi-AI Chat Platform

What it is

LibreChat is a self-hosted AI chat platform that unifies major AI providers in a single privacy-focused interface. It supports Claude (Anthropic), OpenAI, Azure OpenAI, Google AI, and AWS Bedrock. The platform includes AI agents with Model Context Protocol (MCP) support, artifacts, a sandboxed code interpreter for Python, Node.js, and Go, and multi-user authentication.

LibreChat targets teams and individuals who want a unified chat interface across multiple AI providers without sending data through third-party proxy services. With 35,100+ GitHub stars and an MIT license, it has become one of the most popular open-source AI chat frontends.

How it saves time or tokens

Instead of maintaining separate accounts and browser tabs for each AI provider, LibreChat gives you one interface to switch between models mid-conversation. You can compare responses from different providers side by side. The self-hosted deployment means no usage limits imposed by third-party UIs, and you control your API key routing. Built-in conversation history and search eliminate the need for external note-taking.

How to use

1. Clone and start with Docker:

git clone https://github.com/danny-avila/LibreChat.git
cd LibreChat
cp .env.example .env
docker compose up -d

1. Open http://localhost:3080 in your browser and create an account.

1. Add your API keys in the settings panel for each provider you want to use (OpenAI, Anthropic, Google, etc.).

Example

Configuring multiple providers in the .env file:

# OpenAI
OPENAI_API_KEY=sk-...

# Anthropic (Claude)
ANTHROPIC_API_KEY=sk-ant-...

# Google AI
GOOGLE_KEY=...

# Azure OpenAI
AZURE_API_KEY=...
AZURE_OPENAI_BASEURL=https://your-resource.openai.azure.com

After restarting the container, all configured providers appear in the model selector dropdown.

Related on TokRepo

• AI Tools for Coding — developer tools that integrate with AI providers
• AI Gateway Providers — proxy and routing layers for multi-provider AI access

Common pitfalls

• The default Docker Compose uses MongoDB; ensure you have sufficient disk space for conversation storage as usage grows
• MCP server configurations must be added to the LibreChat YAML config, not the provider settings; check the MCP documentation for the correct format
• Multi-user deployments should configure proper auth (LDAP, OIDC) rather than relying on local accounts for team environments