How do I access OpenClaw after ordering?

As soon as the install finishes, an email gives you your Control UI address (on your domain, over HTTPS) and the login password. You open the interface, enter that password, then add your AI provider (OpenAI, Anthropic or local Ollama) to get started. Your server's SSH password is sent separately when the VPS is delivered.

Yes, OpenClaw is open source under the MIT license. You only pay for your VPS and, if you use a remote LLM API (OpenAI/Anthropic), the calls billed by that provider. With a local model via Ollama, there is no API cost at all.

Which AI models can I use?

OpenClaw works with the major LLM providers (OpenAI, Anthropic) or with local models served by Ollama on the same VPS. You choose the provider during onboarding.

What VPS configuration does OpenClaw need?

Count on 2 GB of RAM minimum for the gateway (4 GB is comfortable). If you run a local LLM model via Ollama on the same server, plan for 8 to 16 GB depending on the model.

Are my conversations private?

Yes. The OpenClaw gateway runs on your VPS: your messages pass through your own infrastructure and your keys stay there. With a local model, no data leaves your server.

Build, host and run AI solutions.

OpenClaw

Your personal AI assistant, self-hosted and reachable on your messaging apps — WhatsApp, Telegram, Discord and 50+ channels.

2 GB (8 GB with Ollama) RAM 2 vCPU Port 18789 Available

Source code

Tech stack

DockerNode.js

Minimum RAM2 GB (8 GB with Ollama)

Minimum CPU2 vCPU

Default port18789

Compatible OSubuntu-24.04

OpenClaw (formerly Clawdbot, then Moltbot) is an open-source personal AI assistant that you host yourself. Instead of yet another web interface, it plugs into the messaging apps you already use — WhatsApp, Telegram, Discord, Slack, Signal, Matrix… — and replies there like a correspondent. Created by Peter Steinberger, it became in 2026 one of the most-followed self-hosted projects on GitHub.

Deployed on your VPS, OpenClaw keeps your conversations and your keys under your control: the gateway runs on your own server, you connect the LLM provider of your choice (OpenAI, Anthropic, or a local model via Ollama) and you decide which channels and which skills to enable. It's the sovereign alternative to cloud AI assistants, with no dollar subscription and no data entrusted to a third party.

Key features

Reachable on 50+ messaging apps: WhatsApp, Telegram, Discord, Slack, Signal, Matrix, iMessage…

LLM provider of your choice: OpenAI, Anthropic, or local models via Ollama

Extensible skills and MCP tools

Self-hosted gateway: your conversations and your keys stay on your VPS

Official Docker image (ghcr.io) — simple deployment and updates

Open source (MIT), very active community

When to use this solution?

1

Multi-channel personal assistant

Ask your questions, launch tasks and receive reminders from the messaging app you already have at hand, without installing yet another client.

2

Team concierge

Plug OpenClaw into your team's Slack or Discord to automate replies, summaries and actions across your tools.

3

Sovereign AI

Keep full control: a local model via Ollama, no data sent to a cloud service, zero USD subscription.

Deploy OpenClaw on your VPS

Guide optimized for ServOrbit Cloud VPS.

01

Order OpenClaw

Pick OpenClaw from the catalog and a domain: we provision the VPS and install everything automatically (Ubuntu, Docker, the official ghcr.io/openclaw/openclaw image, HTTPS/Let's Encrypt on your domain).

02

Get your access by email

Once the install completes, you receive your Control UI address (your domain, over HTTPS) and the login password.

03

Log in

Open the Control UI and enter the password from the email.

04

Plug in your AI

From the interface, add your LLM provider: OpenAI, Anthropic, or a local model via Ollama.

05

Connect your messaging apps

Link the channels you want (WhatsApp, Telegram, Discord…) by following the wizard for each integration.

Frequently asked questions

OpenClaw is an open-source (MIT) personal AI assistant that you self-host. It plugs into your messaging apps (WhatsApp, Telegram, Discord and 50+ others) and replies there like a correspondent. Formerly known under the names Clawdbot and Moltbot.

Complementary solutions

Artificial Intelligence

Activate OpenClaw on your infrastructure.

Dedicated Cloud VPS — IPv4 included, European datacenter, support included. Your data never leaves your server.

Recommended configuration: 2 GB (8 GB with Ollama) RAM · 2 vCPU