Instance Overview

What Is an Instance?

An instance is a dedicated OpenClaw server running on ClawHosters. It is your own isolated AI assistant environment with its own configuration, messaging channels, and data.

Each instance runs inside a Docker container on a Hetzner Cloud VPS in Germany. You get full control over which LLM provider it uses, which messaging platforms it connects to, and how it behaves.

Instance Lifecycle

Every instance goes through a series of states from creation to deletion. Here is what each status means:

Normal Flow

The typical path for a new instance:

Provisioning → Deploying → Running

If you chose daily billing without LLM credentials configured:

Provisioning → Deploying → Pending Configuration → Running

Pause and Resume

When your Claws balance gets too low, the system takes a snapshot of your instance and shuts down the server. This is the "paused" state. No billing occurs while paused.

When you add more Claws, you can resume. The system restores the snapshot to a new server and your instance continues where it left off, with all configuration and data intact.

Running → (low balance) → Paused → (add Claws) → Provisioning → Deploying → Running

What Runs Inside an Instance

Each instance contains:

• OpenClaw Gateway — The web interface and API endpoint for your AI assistant
• Messaging adapters — Connectors for Telegram, WhatsApp, Discord, and Slack
• LLM router — Sends prompts to your configured AI provider (Anthropic, OpenAI, Google, etc.)
• Skills and plugins — Tools your AI can use (web browsing, file handling, etc.)
• Configuration files — Your instance settings, channel tokens, and preferences

All of this runs inside a single Docker container on your dedicated VPS. Your data stays on that server and is not shared with other customers.

Instance Dashboard

Each instance has its own dashboard page showing:

• Status — Current state (running, paused, etc.)
• Gateway URL — Direct link to the OpenClaw web interface
• Connected channels — Which messaging platforms are configured
• Resource usage — CPU, memory, and storage consumption
• Quick actions — Start, stop, restart, configure channels, view credentials

Creating an Instance

You create instances through the setup wizard on the ClawHosters dashboard. The wizard walks you through:

1. Billing mode — Monthly subscription or daily Claws
2. Tier selection — Budget, Balanced, or Pro
3. Messaging channels — Configure Telegram, Discord, Slack (WhatsApp is configured after deployment)
4. LLM credentials — Your API key for the AI provider
5. Review and deploy — Confirm settings and start provisioning

The entire process takes about 2-3 minutes from start to a running instance.

Limits and Resources

Each instance is isolated in its own Docker container with resource limits based on the tier:

If the container exceeds its memory limit, Docker restarts it automatically. This causes a brief interruption (a few seconds) but no data loss.

Multiple Instances

You can run multiple instances on your ClawHosters account. Each instance is independent with its own server, configuration, and billing. Common reasons to run multiple instances:

• Different AI assistants for different purposes
• Separate testing and production environments
• Different LLM providers for different use cases

Each instance incurs its own daily Claws cost or monthly subscription charge.

Related Docs

• Creating Your First Instance — Step-by-step creation walkthrough
• Starting and Stopping Instances — How to start, stop, restart, pause, and resume
• Choosing the Right Tier — Tier comparison and selection guide
• Quickstart Guide — Full setup from account creation to running instance