Self-Hosted Deployment

OpenClaw on Custom Server

Deploy OpenClaw on any server you control via DeployClaw

Already have a server, a spare machine, or dedicated hardware? Connect any Linux server to DeployClaw and we install OpenClaw. No specific cloud provider required. This option works for on-premise hardware, colocation servers, or any hosting provider not listed above.

Setup Steps

1Ensure your server runs Ubuntu 22.04+, Debian 12+, or a compatible Linux distribution
2Verify minimum specs: 2 vCPU, 2 GB RAM, 20 GB storage
3Ensure SSH access is available with root or sudo permissions
4Enter your server details in the DeployClaw dashboard

Recommended Specs

Any Linux server with SSH access
Bare metal, VPS, or dedicated hosting
On-premise hardware supported
GPU optional — enables local AI model execution

Starting from $0 server cost + $29/mo DeployClaw

Frequently Asked Questions

What Linux distributions are supported?

Ubuntu 22.04+ and Debian 12+ are officially supported and tested. Other distributions may work but are not guaranteed. We recommend Ubuntu 22.04 LTS for the best compatibility.

Can I use a Raspberry Pi?

A Raspberry Pi 4 or 5 with 4+ GB RAM can run OpenClaw for basic tasks. Performance will be limited compared to cloud servers, especially for tasks involving large file processing or browser automation. ARM64 is supported.

Does my server need a static IP?

A static IP is recommended but not required. If your IP changes, you can update it in the DeployClaw dashboard. For dynamic IPs, consider setting up a dynamic DNS service to maintain consistent connectivity.

Can I run OpenClaw behind a firewall?

Yes. OpenClaw initiates outbound connections for AI provider APIs and messaging platforms. As long as your server has outbound internet access and SSH is reachable by DeployClaw, it works behind corporate firewalls and NATs.