Skip to content

OpenClaw

OpenClaw is a self-hosted gateway that bridges messaging platforms — WhatsApp, Telegram, Slack, Discord, and more — to an LLM. Configure Routeplane as its model provider and every channel routes across the whole registry.

Prerequisites

Routeplane running — local proxy at http://127.0.0.1:4356. Routeplane Cloud is on the Phase D roadmap and not yet shipping; for now, self-host with your own provider keys.
OpenClaw installed. See the OpenClaw docs for setup.

Point OpenClaw at Routeplane

Routeplane is an OpenAI-compatible endpoint, so it goes straight into your OpenClaw config file as a provider — no plugin required. Add a routeplane entry under models.providers and reference it as the default model:

{
  agents: {
    defaults: {
      model: { primary: "routeplane/openai/gpt-4o" },
    },
  },
  models: {
    providers: {
      routeplane: {
        baseUrl: "http://127.0.0.1:4356/v1",
        apiKey: "local-placeholder",   // any value for the local proxy
        api: "openai-completions",
        models: [{ id: "openai/gpt-4o", contextWindow: 128000, maxTokens: 8192 }],
      },
    },
  },
}

Then start the gateway:

openclaw gateway

**No key for the local proxy** — `apiKey` can be any placeholder (the loopback proxy ignores it). Routeplane Cloud is on the Phase D roadmap; the cloud endpoint is not yet shipping.

Pick a model

Each provider model id is a registry id in provider/model form (openai/gpt-4o, anthropic/claude-sonnet-4-6, …), optionally with a :cost / :latency variant. In OpenClaw’s model.primary it’s prefixed with the provider id you chose — e.g. routeplane/openai/gpt-4o. See Models.

Learn more