Integrations/OpenClaw
🦞

OpenClaw

AI Tool

OpenClaw is a personal AI assistant that runs on your devices and chats through your channels. Register Qubax as a custom provider in the OpenClaw gateway config and every agent session can use any of 340+ models.

Official docs

Config (~/.openclaw/openclaw.json)

JSON
{
  "agents": {
    "defaults": {
      "model": { "primary": "qubax/gpt-5" }
    }
  },
  "models": {
    "mode": "merge",
    "providers": {
      "qubax": {
        "baseUrl": "https://api.qubax.ai/v1",
        "apiKey": "YOUR_API_KEY",
        "api": "openai-completions",
        "models": [{ "id": "gpt-5", "name": "GPT-5" }]
      }
    }
  }
}

Setup steps

  1. 1

    Install OpenClaw

    Install OpenClaw following the instructions at openclaw.ai — typically a single install command per platform.

  2. 2

    Open the gateway config

    Edit ~/.openclaw/openclaw.json (the gateway watches this file and hot-reloads changes).

  3. 3

    Register the Qubax provider

    Add the models.providers.qubax block from above: baseUrl set to the Qubax endpoint, your apiKey, api "openai-completions", and at least one model entry.

  4. 4

    Set your primary model

    Set agents.defaults.model.primary to qubax/gpt-5. Pick any model from the Qubax list — swap the id in both the provider models[] array and the primary model ref.

ℹ️
Model references use provider/model syntax — qubax/gpt-5, qubax/claude-opus-5, qubax/glm-5.2. Find all ids on the Qubax Models page or via GET /v1/models.
⚠️
Keep ~/.openclaw/openclaw.json private — it contains your API key.

Test your connection

Paste your Qubax API key and send a live request to verify everything is wired up. Don't have a key yet?

Create one in the dashboard