How to Deploy OpenClaw on Hostinger VPS (Complete Guide)

OpenClaw is an open-source AI agent framework that lets you run personal AI assistants 24/7 on your own infrastructure. Unlike cloud-only solutions (ChatGPT, Claude API), OpenClaw gives you full control over your AI agent — your data stays on your server, you choose the AI model, and you can extend it with custom tools and integrations.

Key features: multi-model support (Claude, GPT, Gemini, Grok), persistent memory across sessions, tool integration (browser, email, calendar, messaging), and a gateway architecture that handles multiple agent instances.

OpenClaw's minimum requirements are modest: 2GB RAM, 1 vCPU, and Docker. Here's why Hostinger is the best fit:

• Price: $4.99/mo for 4GB RAM gives 2x headroom over minimum requirements
• NVMe storage: Fast disk I/O for agent memory and tool caching
• KVM: Docker works perfectly on KVM (unlike some OpenVZ providers)
• Management panel: Easy server monitoring and firewall configuration
• 8 locations: Pick the data center closest to you for low-latency agent interaction

Compared to running OpenClaw on Railway ($5/mo with cold starts) or DigitalOcean ($6/mo for 1GB RAM), Hostinger gives you the most resources for the least money.

Sign up for Hostinger and purchase the KVM 1 plan ($4.99/mo):

1. Go to Hostinger VPS hosting page
2. Select KVM 1 (1 vCPU, 4GB RAM, 50GB NVMe)
3. Choose Ubuntu 22.04 LTS as your operating system
4. Select the data center closest to you
5. Set a strong root password
6. Complete payment and wait 1-2 minutes for provisioning

SSH into your new server and prepare it:

# Connect to your VPS
ssh root@YOUR_SERVER_IP

# Update packages
apt update && apt upgrade -y

# Set up firewall
ufw allow 22/tcp
ufw allow 80/tcp
ufw allow 443/tcp
ufw enable

# Install Docker
curl -fsSL https://get.docker.com | sh
apt install docker-compose-plugin -y

# Create a non-root user
adduser openclaw
usermod -aG sudo,docker openclaw
rsync --archive --chown=openclaw:openclaw ~/.ssh /home/openclaw

# Switch to the new user
su - openclaw

Create the OpenClaw deployment:

# Create project directory
mkdir -p ~/openclaw && cd ~/openclaw

# Create docker-compose.yml
cat > docker-compose.yml << 'EOF'
services:
  openclaw:
    image: openclaw/gateway:latest
    ports:
      - "3000:3000"
    volumes:
      - ./data:/app/data
      - ./config:/app/config
    environment:
      - ANTHROPIC_API_KEY=your-key-here
      - OPENAI_API_KEY=your-key-here
    restart: unless-stopped
EOF

# Create config directory
mkdir -p config data

# Start OpenClaw
docker compose up -d

# Check logs
docker compose logs -f

OpenClaw should be running on port 3000. Visit http://YOUR_SERVER_IP:3000 to verify.

For production use, add HTTPS with Traefik reverse proxy:

# Update docker-compose.yml to use Traefik
cat > docker-compose.yml << 'EOF'
services:
  traefik:
    image: traefik:v3.0
    command:
      - --providers.docker
      - --entrypoints.web.address=:80
      - --entrypoints.websecure.address=:443
      - --certificatesresolvers.letsencrypt.acme.email=your@email.com
      - --certificatesresolvers.letsencrypt.acme.storage=/acme.json
      - --certificatesresolvers.letsencrypt.acme.httpchallenge.entrypoint=web
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
      - ./acme.json:/acme.json
    restart: unless-stopped

  openclaw:
    image: openclaw/gateway:latest
    volumes:
      - ./data:/app/data
      - ./config:/app/config
    environment:
      - ANTHROPIC_API_KEY=your-key-here
    labels:
      - traefik.http.routers.openclaw.rule=Host(`ai.yourdomain.com`)
      - traefik.http.routers.openclaw.tls.certresolver=letsencrypt
    restart: unless-stopped
EOF

# Create acme.json with correct permissions
touch acme.json && chmod 600 acme.json

# Restart
docker compose up -d

Point your domain's DNS to your VPS IP, and Traefik will automatically obtain and renew SSL certificates from Let's Encrypt.

OpenClaw resource usage on Hostinger KVM 1 (4GB RAM):

• Idle gateway: ~200MB RAM, <1% CPU
• Active agent session: ~500MB-1GB RAM depending on context window and tools
• With Traefik + OpenClaw: ~350MB base, leaving 3.6GB for agent operations

The KVM 1 plan handles a single active agent session comfortably. For multi-agent setups (running agents on Discord, Telegram, and WhatsApp simultaneously), upgrade to KVM 2 ($6.99/mo, 8GB RAM).