Installation

Choose your installation method based on your needs:

Development Mode - Fast iteration with hot reload (recommended for development)
Docker - Quick deployment with one command
Docker + HTTPS - Production deployment with automatic SSL certificates

Prerequisites

Choose based on your installation method:

Development Mode:

Bun (recommended) or Node.js 20+
Git

Docker:

Docker and Docker Compose
Git

For Offline AI (optional):

Ollama - For running local models
At least 8GB RAM recommended for local models

Development Mode

Perfect for contributing, customizing, or exploring the codebase.

1. Clone and Install

git clone https://github.com/1337hero/faster-chat.git
cd faster-chat
bun install  # or npm install

2. Start Development Servers

bun run dev  # Starts both frontend and backend

On first run, the server will automatically:

✅ Generate a secure encryption key for API key storage (server/.env)
✅ Create required data directories
✅ Initialize the SQLite database

The application will be available at:

Frontend: http://localhost:3000
API Server: http://localhost:3001

3. First Account Setup

Navigate to http://localhost:3000/login
Click “Register” to create an account
Your first account is automatically promoted to admin

Docker Deployment

Quick deployment with a single command.

1. Clone the Repository

git clone https://github.com/1337hero/faster-chat.git
cd faster-chat

2. Generate Encryption Key

echo "API_KEY_ENCRYPTION_KEY=$(node -e "console.log(require('crypto').randomBytes(32).toString('hex'))")" > server/.env

3. Start with Docker

docker compose up -d

Access at: http://localhost:8787

Docker Configuration

Default Settings:

Port: 8787 (configurable via APP_PORT in server/.env)
Storage: SQLite database in chat-data volume
Runtime: Node.js 22 on Debian (native module compatibility)

Environment Variables (server/.env):

# Required: Encryption key for API keys
API_KEY_ENCRYPTION_KEY=...  # Auto-generated above

# Optional: Configure port
APP_PORT=8787              # Internal port (default: 8787)

# For local Ollama access from Docker
OLLAMA_BASE_URL=http://host.docker.internal:11434

Useful Commands:

docker compose up -d                # Start
docker compose logs -f              # View logs
docker compose down                 # Stop
docker compose up -d --build        # Rebuild

# Reset database
docker compose down
docker volume rm faster-chat_chat-data
docker compose up -d

Docker + HTTPS (Production)

Production deployment with automatic HTTPS certificates via Caddy.

1. Configure Domain

Edit Caddyfile and replace localhost with your domain:

yourdomain.com {
    reverse_proxy app:8787
}

2. Point DNS

Create an A record pointing your domain to your server’s IP address.

3. Start with Caddy

docker compose -f docker-compose.yml -f docker-compose.caddy.yml up -d

Caddy Features:

Automatic HTTPS with Let’s Encrypt
Certificate auto-renewal
HTTP/2 and HTTP/3 support
Compression and security headers
Only 13MB overhead (Alpine-based)

Access at: https://yourdomain.com

For local testing, Caddy also works on localhost without HTTPS:

# Use localhost in Caddyfile
docker compose -f docker-compose.yml -f docker-compose.caddy.yml up -d
# Access at http://localhost

Setting Up Ollama (Optional)

Run AI models completely offline on your local machine.

Install Ollama

macOS / Linux:

curl -fsSL https://ollama.ai/install.sh | sh

Windows: Download from ollama.ai

Pull a Model

# Fast, general-purpose model
ollama pull llama3.2

# Larger, more capable model
ollama pull llama3.1:70b

Start Ollama

ollama serve  # Usually runs automatically

Ollama runs on http://localhost:11434 by default.

Next Steps

Quick Start Guide - Configure providers and start chatting
Provider Configuration - Add OpenAI, Anthropic, or other cloud providers
Admin Guide - Manage users and settings