Installation

Installation

OBEY API Gateway offers multiple deployment methods. Pick the one that matches your environment.

---

Option 1: Download (Windows)

The easiest way to get started on Windows.

1. Download the latest release — choose the installer (.exe) or the portable zip
2. Run the installer or extract the zip to your preferred location
3. Double-click ai-gateway.exe

The gateway starts on http://localhost:8080 and opens the dashboard automatically on first launch.

Windows System Tray Mode

The release build includes system tray integration:

• Splash screen on first launch
• System tray icon with context menu
• Single-instance enforcement (second launch brings existing instance to front)
• Notification when already running

Running as a Windows Service

For always-on operation, register as a Windows service:

Using NSSM (recommended):

# Install NSSM from https://nssm.cc
nssm install ai-gateway "C:\path\to\ai-gateway.exe"
nssm set ai-gateway AppDirectory "C:\path\to"
nssm set ai-gateway AppEnvironmentExtra "OPENAI_API_KEY=sk-..."
nssm start ai-gateway

Using sc.exe:

sc create ai-gateway binPath="C:\path\to\ai-gateway.exe"

---

Option 2: Deploy to Railway

One-click cloud deployment:

Railway picks up the included Dockerfile and railway.toml automatically.

Steps:

1. Click the deploy button above
2. Set your provider API keys (OPENAI_API_KEY, etc.) as environment variables in the Railway dashboard
3. You're live in under a minute

Persist your keys on Railway: Attach a Railway Volume mounted at /data (the image's AI_GATEWAY_DATA_DIR). Railway's container filesystem is ephemeral — without a volume the encryption master key regenerates on every redeploy and previously saved api_key_encrypted values become undecryptable. Alternatively, supply keys via plain environment variables which never touch the encrypted store.

---

Option 3: Docker

Build and Run

# Build the image
docker build -t obey-api-gateway .

# Run with config and env vars
docker run -d --name obey-api-gateway \
  -p 8080:8080 \
  -e OPENAI_API_KEY=sk-... \
  -v $(pwd)/config.yaml:/app/config.yaml \
  -v ai-gateway-data:/data \
  obey-api-gateway --config /app/config.yaml

Persist your keys: The image sets AI_GATEWAY_DATA_DIR=/data and declares it as a volume. Mount a named volume (or host path) at /data so the encryption master key survives container restarts and rebuilds.

Docker Compose

# docker-compose.yml
services:
  obey-api-gateway:
    build: .
    ports:
      - "8080:8080"
    volumes:
      - ./config.yaml:/app/config.yaml
      - ai-gateway-data:/data
    environment:
      - OPENAI_API_KEY=sk-...

volumes:
  ai-gateway-data:

docker compose up -d

Updating (Docker)

# Pull latest source and rebuild
git pull origin master
docker build -t obey-api-gateway .

# Stop and remove old container (data volume is preserved)
docker stop obey-api-gateway && docker rm obey-api-gateway

# Start with new image
docker run -d --name obey-api-gateway \
  -p 8080:8080 \
  -e OPENAI_API_KEY=sk-... \
  -v $(pwd)/config.yaml:/app/config.yaml \
  -v ai-gateway-data:/data \
  obey-api-gateway --config /app/config.yaml

With Compose:

git pull origin master
docker compose up -d --build

Never docker volume rm ai-gateway-data unless you intend to reset all stored secrets.

Exposed Ports

Port	Purpose
8080	Main gateway + admin + dashboard
1455	OAuth callback server (for browser-based OpenAI login)

---

Option 4: Build from Source

Prerequisites

• Rust (stable toolchain)
• Windows, Linux, or macOS

Build

# Clone
git clone https://github.com/fdanobey/OBEY-api-gateway.git
cd OBEY-api-gateway

# Build (headless — no tray icon)
cargo build --release -p ai-gateway

# Build with Windows tray support
cargo build --release -p ai-gateway --features tray

Run

./target/release/ai-gateway --config ./config.yaml

On first run without a config file, a default config.yaml is created automatically.

---

Verifying the Installation

After starting the gateway, verify it's running:

# Health check
curl http://localhost:8080/health

# Check available models
curl http://localhost:8080/v1/models

Open the dashboard in your browser:

http://localhost:8080/dashboard

---

Pointing Your App at the Gateway

Any OpenAI-compatible SDK or tool works:

# Environment variable
export OPENAI_API_BASE=http://localhost:8080/v1

# Python
from openai import OpenAI
client = OpenAI(base_url="http://localhost:8080/v1", api_key="unused")

// TypeScript / Node.js
import OpenAI from 'openai';
const client = new OpenAI({
  baseURL: 'http://localhost:8080/v1',
  apiKey: 'unused'
});

# curl
curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-4-group","messages":[{"role":"user","content":"Hello!"}]}'

---

Next Steps

• Configuration — customize the config file
• Providers — add your AI providers
• Admin Panel & Dashboard — explore the web UI