LibreChat

Configure LibreChat, the open-source ChatGPT alternative, to use Eden AI for access to 200+ AI models.

Overview

LibreChat is a free, open-source AI chat platform that supports multiple providers. By connecting it to Eden AI, you get:

200+ models: Access OpenAI, Anthropic, Google, Cohere, and more through one interface
Self-hosted: Full control over your data and infrastructure
Cost savings: Leverage Eden AI’s competitive pricing
Unified experience: Single chat interface for all providers

Prerequisites

Docker and Docker Compose installed
Eden AI API key from https://app.edenai.run
Basic knowledge of environment variables

Installation

Option 1: Docker Compose (Recommended)

Clone LibreChat and set up with Docker:

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

# Copy example environment file
cp .env.example .env

# Start LibreChat
docker compose up -d

Option 2: Manual Installation

# Clone repository
git clone https://github.com/danny-avila/LibreChat.git
cd LibreChat

# Install dependencies
npm install

# Set up environment
cp .env.example .env

# Build and start
npm run build
npm run backend

Configuration

Step 1: Configure Environment Variables

Edit the .env file to add Eden AI configuration:

# Eden AI Configuration
OPENAI_API_KEY=YOUR_EDEN_AI_API_KEY
OPENAI_REVERSE_PROXY=https://api.edenai.run/v3/llm

# Optional: Enable multiple endpoints
ANTHROPIC_API_KEY=YOUR_EDEN_AI_API_KEY
ANTHROPIC_REVERSE_PROXY=https://api.edenai.run/v3/llm

GOOGLE_API_KEY=YOUR_EDEN_AI_API_KEY
GOOGLE_REVERSE_PROXY=https://api.edenai.run/v3/llm

# Application settings
HOST=localhost
PORT=3080
MONGO_URI=mongodb://127.0.0.1:27017/LibreChat

# Session secret (generate your own)
SESSION_SECRET=your_random_secret_here
JWT_SECRET=your_jwt_secret_here
JWT_REFRESH_SECRET=your_refresh_secret_here

Step 2: Configure librechat.yaml

Create or edit librechat.yaml for advanced configuration:

version: 1.0.5
cache: true

endpoints:
  custom:
    - name: "Eden AI"
      apiKey: "${OPENAI_API_KEY}"
      baseURL: "https://api.edenai.run/v3/llm"
      models:
        default:
          - "anthropic/claude-3-5-sonnet-20241022"
          - "openai/gpt-4"
          - "openai/gpt-4-turbo"
          - "google/gemini-pro"
          - "google/gemini-1.5-pro"
          - "cohere/command-r-plus"
          - "openai/gpt-3.5-turbo"
        fetch: false
      titleConvo: true
      titleModel: "openai/gpt-3.5-turbo"
      summarize: false
      summaryModel: "openai/gpt-3.5-turbo"
      forcePrompt: false
      modelDisplayLabel: "Eden AI"

Step 3: Start LibreChat

# With Docker
docker compose up -d

# Without Docker
npm run backend

Access LibreChat at http://localhost:3080

Available Models

Configure which models appear in the LibreChat interface:

endpoints:
  custom:
    - name: "Eden AI - Claude"
      apiKey: "${OPENAI_API_KEY}"
      baseURL: "https://api.edenai.run/v3/llm"
      models:
        default:
          - "anthropic/claude-3-5-sonnet-20241022"
          - "anthropic/claude-3-opus-20240229"
          - "anthropic/claude-3-sonnet-20240229"
          - "anthropic/claude-3-haiku-20240307"
      modelDisplayLabel: "Claude (Eden AI)"

    - name: "Eden AI - OpenAI"
      apiKey: "${OPENAI_API_KEY}"
      baseURL: "https://api.edenai.run/v3/llm"
      models:
        default:
          - "openai/gpt-4"
          - "openai/gpt-4-turbo"
          - "openai/gpt-4o"
          - "openai/gpt-3.5-turbo"
      modelDisplayLabel: "OpenAI (Eden AI)"

    - name: "Eden AI - Google"
      apiKey: "${OPENAI_API_KEY}"
      baseURL: "https://api.edenai.run/v3/llm"
      models:
        default:
          - "google/gemini-pro"
          - "google/gemini-1.5-pro"
          - "google/gemini-1.5-flash"
      modelDisplayLabel: "Google (Eden AI)"

Features

Multi-Model Conversations

Switch between models mid-conversation:

Start a conversation with Claude
Click the model selector
Switch to GPT-4 or Gemini
Continue the conversation seamlessly

File Attachments

Upload files for vision-capable models:

endpoints:
  custom:
    - name: "Eden AI - Vision"
      apiKey: "${OPENAI_API_KEY}"
      baseURL: "https://api.edenai.run/v3/llm"
      models:
        default:
          - "openai/gpt-4o"
          - "google/gemini-1.5-pro"
          - "anthropic/claude-3-5-sonnet-20241022"
      # Enable file uploads
      fileConfig:
        endpoints:
          assistants:
            fileLimit: 5
            fileSizeLimit: 10
            totalSizeLimit: 50
            supportedMimeTypes:
              - "image/jpeg"
              - "image/png"
              - "image/webp"
              - "image/gif"
      modelDisplayLabel: "Vision Models"

Preset Prompts

Create custom prompts for common tasks:

endpoints:
  custom:
    - name: "Eden AI"
      apiKey: "${OPENAI_API_KEY}"
      baseURL: "https://api.edenai.run/v3/llm"
      models:
        default:
          - "anthropic/claude-3-5-sonnet-20241022"
          - "openai/gpt-4"
      # Add custom presets
      presets:
        - title: "Code Assistant"
          model: "anthropic/claude-3-5-sonnet-20241022"
          temperature: 0.1
          system_message: "You are an expert programmer. Provide clean, well-documented code."

        - title: "Creative Writer"
          model: "openai/gpt-4"
          temperature: 0.9
          system_message: "You are a creative writer. Be imaginative and engaging."

        - title: "Data Analyst"
          model: "google/gemini-1.5-pro"
          temperature: 0.3
          system_message: "You are a data analyst. Provide clear, data-driven insights."

Advanced Configuration

Custom Model Parameters

Configure temperature, max tokens, and other parameters:

endpoints:
  custom:
    - name: "Eden AI"
      apiKey: "${OPENAI_API_KEY}"
      baseURL: "https://api.edenai.run/v3/llm"
      models:
        default:
          - "anthropic/claude-3-5-sonnet-20241022"
      # Default parameters
      default:
        temperature: 0.7
        max_tokens: 2000
        top_p: 1.0
        frequency_penalty: 0.0
        presence_penalty: 0.0
      # Allow users to override
      userProvidedParameters:
        - "temperature"
        - "max_tokens"
        - "top_p"

User Authentication

Enable user registration and authentication:

# Allow user registration
ALLOW_REGISTRATION=true

# Email verification (optional)
EMAIL_SERVICE=gmail
EMAIL_USERNAME=[email protected]
EMAIL_PASSWORD=your-app-password

# Social auth (optional)
GOOGLE_CLIENT_ID=your-google-client-id
GOOGLE_CLIENT_SECRET=your-google-client-secret
GOOGLE_CALLBACK_URL=http://localhost:3080/oauth/google/callback

Rate Limiting

Protect your API key with rate limiting:

rateLimits:
  fileUploads:
    ipMax: 100
    ipWindowInMinutes: 60
    userMax: 50
    userWindowInMinutes: 60

  conversationsImport:
    ipMax: 10
    ipWindowInMinutes: 60
    userMax: 5
    userWindowInMinutes: 60

Conversation History

Configure MongoDB for persistent conversations:

# MongoDB connection
MONGO_URI=mongodb://127.0.0.1:27017/LibreChat

# Or use MongoDB Atlas (cloud)
# MONGO_URI=mongodb+srv://username:[email protected]/LibreChat

Docker Deployment

Production Docker Compose

version: '3.4'

services:
  api:
    image: ghcr.io/danny-avila/librechat:latest
    container_name: LibreChat
    ports:
      - 3080:3080
    depends_on:
      - mongodb
    restart: always
    environment:
      - HOST=0.0.0.0
      - MONGO_URI=mongodb://mongodb:27017/LibreChat
      - OPENAI_API_KEY=${OPENAI_API_KEY}
      - OPENAI_REVERSE_PROXY=https://api.edenai.run/v3/llm
    volumes:
      - ./librechat.yaml:/app/librechat.yaml
      - ./images:/app/client/public/images

  mongodb:
    image: mongo
    container_name: chat-mongodb
    restart: always
    volumes:
      - ./data-node:/data/db
    ports:
      - 27017:27017

Deploy to Production

# Pull latest image
docker compose pull

# Start services
docker compose up -d

# View logs
docker compose logs -f api

# Check status
docker compose ps

Troubleshooting

Models Not Appearing

If models don’t show up in the interface:

Check librechat.yaml syntax:
```
# Validate YAML
docker compose config
```

Verify API key:

# Test Eden AI endpoint
curl -X POST https://api.edenai.run/v3/llm/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-3-5-sonnet-20241022",
    "messages": [{"role": "user", "content": "test"}],
    "stream": true
  }'

Clear cache and restart:

docker compose down
docker compose up -d

Authentication Errors

If you see 401 errors:

Check .env has correct API key
Ensure no extra spaces in the key
Verify OPENAI_REVERSE_PROXY URL is correct
Restart services after changing .env

Slow Responses

If responses are slow:

Use faster models for chat titles:
```
titleModel: "openai/gpt-3.5-turbo"
```
Disable unnecessary features:
```
summarize: false
```
Check your internet connection and Eden AI status

Connection Refused

If LibreChat can’t connect to MongoDB:

Check MongoDB is running:
```
docker compose ps mongodb
```

Verify MONGO_URI in .env:

MONGO_URI=mongodb://mongodb:27017/LibreChat

Check network connectivity:
```
docker compose logs mongodb
```

Security Best Practices

1. Secure API Keys

Never commit API keys to version control:

2. Use Environment-Specific Configs

# Production settings
NODE_ENV=production
ALLOW_REGISTRATION=false
SESSION_SECRET=strong_random_secret_here

3. Enable HTTPS

Use a reverse proxy like Nginx:

server {
  listen 443 ssl;
  server_name chat.yourdomain.com;

  ssl_certificate /path/to/cert.pem;
  ssl_certificate_key /path/to/key.pem;

  location / {
    proxy_pass http://localhost:3080;
    proxy_http_version 1.1;
    proxy_set_header Upgrade $http_upgrade;
    proxy_set_header Connection 'upgrade';
    proxy_set_header Host $host;
    proxy_cache_bypass $http_upgrade;
  }
}

4. Implement Rate Limiting

Protect against abuse:

rateLimits:
  fileUploads:
    ipMax: 50
    ipWindowInMinutes: 60
  conversationsImport:
    ipMax: 5
    ipWindowInMinutes: 60

Cost Optimization

1. Use Appropriate Models

Configure cheaper models for simple tasks:

endpoints:
  custom:
    - name: "Eden AI"
      # Use GPT-3.5 for titles (cheaper)
      titleModel: "openai/gpt-3.5-turbo"
      # Use Claude Haiku for summaries (fast & cheap)
      summaryModel: "anthropic/claude-3-haiku-20240307"

2. Monitor Usage

Track costs through Eden AI dashboard:

View usage at https://app.edenai.run
Set up billing alerts
Review monthly reports

3. Limit Token Usage

endpoints:
  custom:
    - name: "Eden AI"
      default:
        max_tokens: 1000  # Limit response length

Example Use Cases

1. Team Collaboration

Set up LibreChat for your team:

Enable user registration for team members
Configure multiple endpoints for different projects
Use presets for common workflows (code review, documentation, etc.)

2. Customer Support

Deploy as an internal support tool:

Create presets for support responses
Use conversation history to maintain context
Configure rate limits to prevent abuse

3. Development Assistant

Integrate with your development workflow:

Code assistance with Claude or GPT-4
Documentation generation with presets
Bug analysis with vision models (screenshots)

Next Steps

Open WebUI - Alternative chat interface
Python SDK - Programmatic access
Cost Management - Track spending

V3 Documentation

​LibreChat

​Overview

​Prerequisites

​Installation

​Option 1: Docker Compose (Recommended)

​Option 2: Manual Installation

​Configuration

​Step 1: Configure Environment Variables

​Step 2: Configure librechat.yaml

​Step 3: Start LibreChat

​Available Models

​Features

​Multi-Model Conversations

​File Attachments

​Preset Prompts

​Advanced Configuration

​Custom Model Parameters

​User Authentication

​Rate Limiting

​Conversation History

​Docker Deployment

​Production Docker Compose

​Deploy to Production

​Troubleshooting

​Models Not Appearing

​Authentication Errors

​Slow Responses

​Connection Refused

​Security Best Practices

​1. Secure API Keys

​2. Use Environment-Specific Configs

​3. Enable HTTPS

​4. Implement Rate Limiting

​Cost Optimization

​1. Use Appropriate Models

​2. Monitor Usage

​3. Limit Token Usage

​Example Use Cases

​1. Team Collaboration

​2. Customer Support

​3. Development Assistant

​Next Steps