Set Up OpenClaw on Vultr

1Create a Vultr account

Plan to use a clean Ubuntu VPS, not a one-click app image.

Official docs list 1 vCPU / 1GB RAM as the absolute minimum and 1-2 vCPU / 2GB+ as a safer starting point for a long-running gateway.

2Deploy a Server

In your Vultr dashboard, deploy a new server:

1. Click Deploy New Server
2. Choose a standard cloud compute plan
3. Select a server location (choose the one closest to your users)
4. Choose Ubuntu 24.04 LTS as the OS
5. Start with a small plan and scale up only if you need more headroom
6. Deploy the server and note its IP address

Vultr is workable, but this page is a provider adaptation rather than an official OpenClaw walkthrough.

3Connect to Your Server

Open a terminal and connect to your server via SSH:

ssh root@YOUR_SERVER_IP

Replace YOUR_SERVER_IP with your actual server IP address. Enter the root password from your Vultr dashboard when prompted.

4Install OpenClaw CLI

Update the machine and install OpenClaw using the official installer:

sudo apt update && sudo apt upgrade -y
curl -fsSL https://openclaw.ai/install.sh | bash
openclaw --version

5Run onboarding

Start the onboarding flow and install the Gateway service:

openclaw onboard --install-daemon

Use onboarding to configure model auth, channels, and the gateway token.

6Open the dashboard securely

Prefer an SSH tunnel or Tailscale instead of exposing the Control UI publicly:

ssh -N -L 18789:127.0.0.1:18789 root@YOUR_SERVER_IP
# Then open http://127.0.0.1:18789/ locally

7Verify the gateway and pair your first channel

Back on the server, confirm the service and reopen the dashboard link when needed:

openclaw status
openclaw dashboard

Use `openclaw status` for a quick health check. If auth is required, paste the token into the Control UI settings.