No need to buy a Mac Mini! Run OpenClaw 24/7 at a low cost of $6/month with Hostinger VPS

I used to assume running OpenClaw long-term meant dedicating a machine to it — either a Mac Mini sitting at home or an expensive cloud server. Turns out that's completely unnecessary. Getting the whole thing down to $6/month on a Hostinger VPS is entirely doable.

You really only need two things: a Hostinger VPS and an OpenRouter API key.

Why Hostinger VPS

The Hostinger KVM2 plan is the foundation of this setup: 2 vCPU cores, 8GB RAM, 100GB NVMe SSD, running around $6/month on a two-year plan.

The 8GB RAM matters — AI agents need enough memory headroom to stay stable, and you'll notice the difference if you're running on less. The NVMe storage is faster than standard SSD, which shows up in deployment and startup times. Hostinger also accepts Alipay, which removes the payment friction for users in China.

Running locally has obvious problems: the agent goes offline when your computer does, your IP isn't stable, and remote management is a hassle. A VPS solves all of that — 24/7 uptime, a dedicated public IP, and you can send instructions via Telegram from your phone without needing a computer open.

Why OpenRouter

One OpenRouter API key gives you access to 400+ models — Minimax, Claude, GPT, and more — with the ability to route between them based on task type. The cost efficiency comes from the model strategy: use cheaper models for routine tasks, only pull in a top-tier model when you actually need deep reasoning. Monthly model costs can stay in the single digits, which is over 20x cheaper than a standalone Claude Opus subscription.

Full Deployment Walkthrough

Step 1: Purchase Hostinger KVM2 VPS

Go to Hostinger, navigate to VPS Hosting, and select the KVM2 plan.

• OS: Ubuntu 22.04
• Plan: KVM2 (2 cores, 8GB RAM)
• Data center: pick the node closest to your target users — Europe for European audiences, Singapore or US West for Asia-Pacific

After purchase, Hostinger will email you the server IP and initial root password. You can also find both in the hPanel dashboard.

Step 2: SSH Into the Server

Open a terminal (Mac/Linux use the system terminal, Windows use PowerShell or PuTTY):

ssh root@your-server-ip

Enter the initial password. Once you're in, update the system:

apt update && apt upgrade -y

For security, create a non-root user rather than operating as root long-term:

adduser yourname
usermod -aG sudo yourname
su - yourname

Step 3: Install Docker

Hostinger's Ubuntu 22.04 image doesn't include Docker by default:

sudo apt install -y ca-certificates curl gnupg
sudo install -m 0755 -d /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg
sudo chmod a+r /etc/apt/keyrings/docker.gpg

echo \
  "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] \
  https://download.docker.com/linux/ubuntu \
  $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | \
  sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

sudo apt update
sudo apt install -y docker-ce docker-ce-cli containerd.io docker-compose-plugin

Verify the installation:

sudo docker --version

Add your user to the docker group so you don't need sudo every time:

sudo usermod -aG docker $USER
newgrp docker

Step 4: Deploy OpenClaw

Pull the image and start the container:

docker pull openclaw/openclaw:latest

docker run -d \
  --name openclaw \
  --restart always \
  -p 8080:8080 \
  -v ~/.openclaw:/app/data \
  openclaw/openclaw:latest

A note on the flags:

• --restart always: the container starts automatically if the VPS reboots — no manual intervention needed
• -p 8080:8080: maps the container's port 8080 to the host
• -v ~/.openclaw:/app/data: persists data to the host filesystem so it survives container removal

Confirm the container is running:

docker ps

If you see openclaw with status Up, the deployment is done.

Step 5: Configure the Hostinger Firewall

Hostinger hPanel has a built-in firewall. Find the Firewall settings in the dashboard and add a rule to open port 8080.

Alternatively, configure ufw directly on the server:

sudo ufw allow 8080
sudo ufw allow ssh
sudo ufw enable

Step 6: Configure OpenRouter

Go to openrouter.ai, create an account, navigate to API Keys, and generate a new key.

Open a browser and go to:

http://your-hostinger-server-ip:8080

In the OpenClaw settings page, fill in:

• API Base URL: https://openrouter.ai/api/v1
• API Key: the key you just generated
• Default model: minimax/minimax-01 (cheap and capable enough for most tasks)

Save and run a connection test to confirm the API is working.

Step 7: Configure Model Routing

In OpenClaw's CloudOps settings, assign models by task type:

Primary Agent (routine tasks)  →  minimax/minimax-01
Sub Agent (data processing)    →  deepseek/deepseek-chat
Complex reasoning              →  anthropic/claude-3-5-sonnet
Fallback                       →  openai/gpt-4o-mini

The vast majority of tasks stay on the cheaper models. Top-tier models only get called when genuinely needed, which keeps monthly model spend low.

Step 8: Connect Telegram

In Telegram, search for BotFather, send /newbot, and follow the prompts to set a bot name. You'll receive a token that looks like:

1234567890:ABCdefGHIjklMNOpqrSTUvwxYZ

Paste the token into the Telegram Bot Token field in OpenClaw settings and save. Find your new bot in Telegram, send /start, and if you get a reply, the connection is live. From here on, you can send instructions to your AI agent directly from your phone — no browser, no dashboard.

Step 9: Set Up Automatic Memory Backup

Agents accumulate memory data over time, and Hostinger VPS occasionally restarts for maintenance. Worth setting up auto-sync to a private GitHub repo.

Create a private repository on GitHub, then generate a Personal Access Token (Settings → Developer settings → Personal access tokens → Generate new token).

Configure Git on the server:

git config --global user.email "[email protected]"
git config --global user.name "yourname"

Create the sync script:

nano ~/.openclaw/sync.sh

Add the following:

#!/bin/bash
cd ~/.openclaw
git init
git remote add origin https://yourname:[email protected]/yourname/openclaw-backup.git 2>/dev/null || true
git add .
git commit -m "sync $(date '+%Y-%m-%d %H:%M:%S')"
git push -u origin main 2>/dev/null || git push origin main

Make it executable:

chmod +x ~/.openclaw/sync.sh

Set up a cron job to sync every 10 minutes:

crontab -e

Add this line at the end of the file:

*/10 * * * * ~/.openclaw/sync.sh >> ~/.openclaw/sync.log 2>&1

Save and exit. Backup is configured.

Cost Comparison

Most people running this in practice end up under $10/month total.

Routine Maintenance Commands

Once it's running, there's very little to do. Commands you'll occasionally need:

Restart the container:

docker restart openclaw

View live logs:

docker logs -f openclaw

Check memory usage:

docker stats openclaw

Update to the latest version:

docker pull openclaw/openclaw:latest
docker stop openclaw
docker rm openclaw
docker run -d --name openclaw --restart always -p 8080:8080 -v ~/.openclaw:/app/data openclaw/openclaw:latest

Who This Is For

If you want an AI agent running 24/7 — for automation, content generation, client deliverables, or just as a reliable long-term tool — this setup covers it. Hostinger KVM2 has more than enough headroom for OpenClaw, the pricing is competitive among similar VPS providers, and Alipay support makes it accessible for users in China. It's one of the better value options available right now for running OpenClaw at this kind of sustained workload.