mcp-server

LibreChat

Name: LibreChat
Author: danny-avila

Enhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching,…

42,054 stars8,697 forksTypeScriptUpdated 8/15/2026100% free · open source

What it does

LibreChat is an enhanced ChatGPT clone that allows for multi-user authentication, AI model switching, and integration with various cloud services and AI models, enabling founders to build custom conversational AI solutions.

Install / run

git clone https://github.com/danny-avila/LibreChat.git && cd LibreChat

When to use it

•When you need a customizable conversational AI platform for your startup
•When you want to integrate multiple AI models and services into a single chat interface
•When you require secure multi-user authentication for your conversational AI application

Quick start

1Run the command `npm install` to install the required dependencies
2Create a new file named `config.json` and configure your AI models and services, such as OpenAI or AWS, by following the examples in the `README.md` file
3Start the server with `npm run start` and access the chat interface at `http://localhost:3000`
4Use the `presets` feature to load pre-configured settings for your AI models and services, such as the `default` preset
5Test the `message search` feature by sending a message and then searching for it using the search bar

Ready-to-paste prompt

Send a message to the chat interface with the text `@openai what is the weather like today?` to use the OpenAI model to respond to the user's question

Heads up: Make sure you have a valid API key for the AI services you want to use, such as OpenAI or AWS, and configure them correctly in the `config.json` file to avoid authentication errors

Saves to your device

Topics

anthropic

artifacts

aws

azure

chatgpt

chatgpt-clone

claude

clone

deepseek

gemini

google

gpt-5

librechat

mcp

openai

responses-api

vision

webui

How LibreChat works

Codeflow

Free to inspect

LibreChat is a very large TypeScript project (~696k lines across 3247 code files, plus 744 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, MIT license, has a test suite.

Size

Very large codebase

~696k lines · 3247 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Node.js · Docker

Needs API keys (.env)

TypeScript 81%JavaScript 19%

Where to start reading