Frequently Asked Questions

Common questions about Clawpy, pricing, and self-hosting.

What LLM providers does Clawpy support?

Clawpy supports all major foundation model providers through a unified routing layer:

You can mix and match providers — use a fast model for triage and a powerful model for complex reasoning, all within the same swarm.

How much does it cost?

Self-hosted Clawpy is free and open source. You only pay your LLM provider for API usage. There are no Clawpy platform fees.

Clawpy uses a Bring Your Own Key (BYOK) model. You configure your own API keys in the Vault, and all costs are billed directly by your provider.

Clawpy Cloud (coming soon) will offer managed hosting with additional features. Pricing will be announced when it launches.

Is my data private?

Yes. In self-hosted mode, all data stays on your infrastructure. Clawpy never phones home or sends telemetry. Your API keys are encrypted at rest using Fernet encryption in the local Vault.

Can I run Clawpy on my laptop?

Yes, but we recommend at least 8GB of RAM and Docker Desktop installed. For production workloads, a VPS with 4+ cores and 16GB RAM is recommended.

Do I need all the LLM API keys?

No. You only need one API key from any supported provider to get started. Additional keys unlock access to more models and the ability to assign different providers to different agents.

Can agents access the internet?

Agents can use built-in tools like Brave Search and web browsing when enabled. All external access is controlled through the skill system — you decide which tools each agent can use.

How is Clawpy different from ChatGPT or Claude?

Single-model assistants run one conversation at a time. Clawpy orchestrates a team of specialised agents that collaborate, delegate, and work autonomously in the background. Each agent has persistent memory, a defined role, and runs in an isolated sandbox.