Simple pricing,
serious infrastructure.

OpenClaw is an open-source AI gateway that lets you manage, route, and monitor LLM API calls across multiple providers. HostedShell provides managed hosting so you can run OpenClaw on Kubernetes without managing infrastructure.

With BYOK, you use your existing API keys from providers like OpenAI, Anthropic, Cohere, etc. We securely store them in an encrypted vault and configure your OpenClaw instance to use them. You pay the providers directly — we just host the infrastructure.

You deposit credits starting at $5. When your OpenClaw instance processes requests, we debit your balance at transparent per-token rates. We add a small margin on top of provider costs, which is clearly displayed in your dashboard. No surprises.

Yes. You can switch plans at any time from your dashboard. Your instances and configuration are preserved. If you switch from BYOK to Credits, we will provision managed API keys for you automatically.

Your instances will gracefully queue requests for a short grace period. We send alerts at 25%, 10%, and 5% remaining balance. You can set up auto-reload to ensure uninterrupted service.

Yes. Enterprise plans include dedicated clusters, custom SLAs, SSO/SAML, audit logs, a dedicated customer success manager, and volume discounts. Contact our sales team for a custom quote.

Most deployments complete in under 30 seconds. We pre-warm Kubernetes clusters and use optimized Helm charts. Your OpenClaw instance is production-ready with SSL, monitoring, and auto-scaling from the start.

OpenClaw supports all major providers: OpenAI (GPT-4, GPT-3.5), Anthropic (Claude), Google (Gemini), Cohere, Mistral, and any OpenAI-compatible API. You can also configure custom endpoints for self-hosted models.