Integration guide

Run OpenClaw on OpenLux models

OpenClaw accepts custom OpenAI-compatible providers through models.providers in ~/.openclaw/openclaw.json. Register OpenLux once and your self-hosted assistant can run any catalogue model — GPT-6 Astra, Claude, Gemini, DeepSeek — on one key.

Create an API key Read the docs

Abstract orange glass sculpture illustrating the OpenClaw integration with OpenLux — OpenClaw + OpenLux — one key, one endpoint, one bill.

Set it up in four steps

Install OpenClaw

Install the gateway (npm install -g openclaw) and finish onboarding; its config lives in ~/.openclaw/openclaw.json.

Create an OpenLux key

Create a key in the console and export it as OPENLUX_API_KEY in the environment that runs the gateway.

Register OpenLux as a provider

Add an openlux entry under models.providers with the base URL, the environment-variable key and the models your assistant should see.

Point the agent at a model

Set agents.defaults.model.primary to an OpenLux model reference and restart the gateway; openclaw models list confirms the provider is live.

Configuration

Copy these blocks, replace the placeholder key and you are done.

~/.openclaw/openclaw.json

{
  agents: {
    defaults: { model: { primary: "openlux/gpt-6-astra" } },
  },
  models: {
    mode: "merge",
    providers: {
      openlux: {
        baseUrl: "https://openlux.lol/v1",
        apiKey: "${OPENLUX_API_KEY}",
        api: "openai-completions",
        models: [
          { id: "gpt-6-astra", name: "GPT-6 Astra" },
          { id: "claude-opus-5", name: "Claude Opus 5" },
          { id: "deepseek-v4", name: "DeepSeek V4" },
          { id: "gemini-3.8-flash", name: "Gemini 3.8 Flash" }
        ]
      }
    }
  }
}

Terminal

export OPENLUX_API_KEY="sk-openlux-..."
openclaw models list

Models that fit OpenClaw

GPT-6 Astra

OpenAI · $0.3677 / 1M tokens

Claude Opus 5

Anthropic · $0.4412 / 1M tokens

DeepSeek V4

DeepSeek · $0.0613 / 1M tokens

Gemini 3.8 Flash

Google · $0.0551 / 1M tokens

The full list is in the model catalogue and the use-case collections.

OpenClaw + OpenLux questions

Does OpenClaw support OpenAI-compatible gateways?

Yes. models.providers takes custom entries with a baseUrl and api: "openai-completions" — the same mechanism the OpenClaw docs use for other compatible providers.

Which OpenLux models suit a personal assistant?

GPT-6 Astra or Claude Opus 5 as the primary brain, Gemini flash tiers for routine messages and DeepSeek V4 when cost matters. Each is one entry away in the models list.

Where should the API key live?

Reference ${OPENLUX_API_KEY} in the config so the file itself holds no secret, then export the variable in the environment that runs the gateway.

Can I run OpenClaw fully self-hosted on OpenLux?

Yes. OpenClaw runs on your hardware and OpenLux is a plain HTTPS API, so the only traffic leaving your machine is the model calls themselves, billed per token.

Connect OpenClaw to OpenLux today

One key covers every model in this guide and the rest of the catalogue, metered in USD with free trial credit to start.

Create an API key See pricing