Deploy OpenClaw on a VPS in 5 Minutes: Your Own AI Assistant, Your Rules

Deploy OpenClaw on a VPS in 5 Minutes: Your Own AI Assistant, Your Rules

OpenClaw is the fastest-growing open source project of 2026, with over 247,000 GitHub stars and counting. It's a self-hosted AI assistant that connects your favorite messaging apps — WhatsApp, Telegram, Discord, Slack, and more — to powerful AI models like Claude and GPT-4. Instead of paying for yet another SaaS subscription, you run it on your own server, keep your data private, and only pay for the API calls you actually make.

In this guide, we'll show you how to deploy OpenClaw on a Dedimax VPS using Docker in under 5 minutes.


Why Self-Host OpenClaw on a VPS?

You can run OpenClaw on your laptop, but the moment you close it, your assistant goes offline. A VPS keeps it running 24/7 without depending on your personal machine.

Beyond uptime, a VPS gives you a stable network connection with a dedicated IP, which is critical for WebSocket-based messaging integrations like WhatsApp and Telegram. It also isolates the agent from your personal files and network, adding a layer of security since OpenClaw can access the filesystem and execute commands.

And the cost? A capable VPS for OpenClaw starts at just $9.99/month. Compare that to a ChatGPT Plus subscription at $20/month — and with OpenClaw you get multi-channel support, full customization, and complete data ownership.


What You Need

OpenClaw is lightweight. Here are the minimum requirements:

  • A Linux VPS with at least 2 vCPUs, 4 GB RAM, and 20 GB SSD (4 GB RAM recommended for stable 24/7 operation)
  • Docker and Docker Compose v2
  • An AI API key (Anthropic Claude recommended, or OpenAI)
  • About 10 minutes of your time

Step 1: Get Your VPS on Dedimax

Head to dedimax.com and pick a VPS. For OpenClaw, our $9.99/month plan with 4 GB RAM and 2 vCPUs is the sweet spot.

Why Dedimax for OpenClaw specifically?

Unlimited bandwidth. OpenClaw maintains persistent WebSocket connections to multiple messaging platforms simultaneously. These connections transfer data continuously. Many hosting providers cap bandwidth at 1-8 TB/month — if you're running an active assistant handling dozens of messages daily across multiple channels, you could hit those limits. With Dedimax, bandwidth is unlimited.

No commitment. Want to try OpenClaw for a month and see if it fits your workflow? Pay for one month, no strings attached. If it's not for you, cancel anytime. No 12 or 24-month contract required.

120+ server locations. Deploy your assistant close to your users for lower latency. Running a team assistant for your European office? Pick our French or German datacenter. Serving customers in Asia? We've got locations there too.

Once your VPS is ordered, you'll receive your SSH credentials within minutes. Choose Ubuntu 22.04 or 24.04 as your operating system.


Step 2: Install Docker

SSH into your VPS:

ssh root@YOUR_VPS_IP

Install Docker using the official convenience script:

curl -fsSL <https://get.docker.com> | sh
systemctl enable docker
docker --version
docker compose version

Both commands should return version numbers. Docker is ready.


Step 3: Deploy OpenClaw

Clone the OpenClaw repository and run the Docker setup script:

git clone <https://github.com/openclaw/openclaw.git>
cd openclaw
./docker-setup.sh

The setup script will build the Docker image and launch an interactive onboarding wizard. It will ask you several questions:

Gateway mode: Choose "Local". This means the gateway runs on your VPS.

AI provider: Select your preferred model provider. Anthropic (Claude) is the recommended default. You'll need to paste your API key — get one from console.anthropic.com if you don't have one yet.

Messaging channels: Connect the platforms you want to use. WhatsApp, Telegram, Discord, and Slack are the most popular. Each channel has its own pairing process — the wizard will guide you through it.

Gateway token: The setup generates a secure token for accessing the web dashboard. Save this token somewhere safe.

Once the wizard completes, OpenClaw starts automatically. That's it — your AI assistant is live.


Step 4: Verify and Access

Check that the containers are running:

docker compose ps

You should see the gateway container with status "Up".

Access the web dashboard at http://YOUR_VPS_IP:18789?token=YOUR_GATEWAY_TOKEN to manage your assistant, view conversations, and configure settings.

To pair a Telegram bot (for example), run:

docker compose run --rm openclaw-cli pairing approve telegram YOUR_CODE

Send a message to your bot on Telegram — OpenClaw will respond using your configured AI model.


Step 5: Secure Your Deployment

Before sharing your assistant with others, lock down the basics.

Configure the firewall to only allow SSH and the OpenClaw ports:

ufw allow 22/tcp
ufw allow 18789/tcp
ufw enable

For production use, set up a reverse proxy with Nginx and a free SSL certificate from Let's Encrypt. This gives you HTTPS access and protects the gateway token in transit:

apt install nginx certbot python3-certbot-nginx -y

Configure Nginx to proxy traffic to OpenClaw on port 18789, then run certbot --nginx to get your SSL certificate. The official OpenClaw documentation has detailed instructions for this step.

Also consider restricting the gateway to loopback mode and accessing it only through SSH tunnels or Tailscale for maximum security. This prevents the dashboard from being exposed to the public internet.


Step 6: Keep It Updated

OpenClaw releases frequently. To update your deployment:

cd openclaw
git pull
docker compose pull
docker compose up -d
docker image prune -f

Your conversations, settings, and connected channels are stored in Docker volumes and persist across updates.


OpenClaw vs. ChatGPT: Why Self-Hosting Wins

ChatGPT is great, but it's a walled garden. OpenClaw gives you something fundamentally different.

With ChatGPT, your conversations live on OpenAI's servers. With OpenClaw on your VPS, everything stays on your machine. No one else can read your messages, your files, or your assistant's memory.

ChatGPT works in a browser tab. OpenClaw works in the apps you already use every day — WhatsApp, Telegram, Discord, Slack, and more. Send a message from your phone and get an AI response in the same chat thread.

ChatGPT Plus costs $20/month for a fixed set of features. With OpenClaw, you pay for the VPS ($9.99/month on Dedimax) plus only the API calls you actually make. For moderate usage, this is significantly cheaper.

And the killer feature: OpenClaw can execute tasks, not just chat. It can manage files, run shell commands, browse the web, and trigger automations. It's not an assistant you talk to — it's an assistant that works for you.


Ready to Deploy?

In under 10 minutes, you can have your own AI assistant running 24/7, connected to all your messaging platforms, with full privacy and control — for less than $10/month.

Head to dedimax.com to get your VPS and start deploying OpenClaw today.

Continue reading

Zona comunitaria

Una pregunta ?
¡Encuentra respuestas y comparte tus conocimientos!

Te estamos esperando zona comunitaria. Más que 70 guías (sysadmin, gaming, devops...) !

Permítame verificar
DEDIMAX DEDIMAX DEDIMAX DEDIMAX
DEDIMAX

¿Necesita una cotización?

Escribenos !

Contáctenos

Prendre contact