Pick a plan. Deploy in minutes.
Your subscription covers the platform — deployment automation, dashboard, updates, and support. Server hosting and AI inference are billed separately by your own providers.
Solo
For individual makers, freelancers, and tinkerers who want a personal AI operator on their own hardware
$23/mo billed annually · + server & AI costs
- 2 agent instances
- 1 team seat
- All messaging platforms
- Unlimited messages
- 5 scheduled automations
- Weekly backup snapshots
- 7-day log retention
- Read-only API
- Community forum support
- Custom domain
- Audit trail
Team
For engineering teams and small agencies that need shared access, daily backups, and serious automation
$39/mo billed annually · + server & AI costs
- 10 agent instances
- 5 team seats
- All messaging platforms
- Unlimited messages
- 50 scheduled automations
- Daily backup snapshots
- 30-day log retention
- Full CRUD API
- Email support (24h response)
- Custom domain with auto-SSL
- Basic audit trail
Business
For organizations with compliance requirements, multi-client operations, and teams larger than five
$119/mo billed annually · + server & AI costs
- Unlimited agent instances
- 25 team seats
- All messaging platforms
- Unlimited messages
- Unlimited scheduled automations
- Hourly snapshots + point-in-time restore
- 90-day log retention + export
- Full CRUD API + webhooks
- Dedicated Slack channel (4h response)
- Custom domain + white-label dashboard
- Full audit trail with compliance export
- SSO / SAML
Server hosting is separate
Your cloud provider bills you directly. A production-ready VPS starts at a few dollars per month.
Common questions
What exactly do I pay for with an OpenClaw subscription?
Your subscription covers the deployment platform: automated installation, the management dashboard, rolling security updates, backup orchestration, and support. AI model inference is billed by your chosen provider (Anthropic, OpenAI, etc.), and server hosting goes through your cloud provider. We never mark up third-party costs.
Can I test OpenClaw before paying?
Every plan includes a 14-day money-back guarantee. Deploy on your server, connect your platforms, and put the agent through its paces. If it doesn't fit, email support for a full refund.
What hardware do I need?
Minimum: 2 vCPU, 2 GB RAM, 20 GB disk on Ubuntu 22.04 or Debian 11+. That runs the agent with a cloud AI provider. If you want fully offline inference with Ollama or llama.cpp, you'll need a GPU with at least 8 GB VRAM.
What is an agent instance?
One instance is a running OpenClaw process on your server. Each instance connects to all your messaging platforms simultaneously and can execute tasks independently. The Solo plan includes two instances; Team allows ten; Business is unlimited.
Can I change plans later?
Upgrades and downgrades take effect immediately with prorated billing. No penalties, no waiting periods.
How does annual billing work?
Paying annually saves roughly 20%. Solo drops to $23/mo ($276/year), Team to $39/mo ($468/year), and Business to $119/mo ($1,428/year). You can switch between monthly and annual at any time.
Is there a long-term contract?
No. Cancel anytime. You keep access through the end of your billing period, and your server and data remain yours.
What payment methods do you accept?
Visa, Mastercard, Amex, and PayPal — processed through Stripe with PCI-compliant infrastructure.
Not sure which plan fits?
Reach out and we will help you find the right tier for your workload.