Configure in OpenClaw
OpenClaw is a powerful AI assistant that supports multiple AI models and interaction channels.
Prerequisites
- You need to install Node.js 22 or higher
Install OpenClaw
Choose any of the following methods to install:
Choose an installation method
curl -fsSL https://openclaw.ai/install.sh | bash
iwr -useb https://openclaw.ai/install.ps1 | iex
Configure OpenClaw
After running the installation command above, the configuration process will start automatically. If it doesn't start, you can run the following command to begin configuration:
openclaw onboard
The wizard will interactively guide you through: AI models, chat channels, skills, hooks, gateway services, etc.
- Security Notice: Read and select Yes to continue
- Onboarding Mode: Select QuickStart (recommended), using default port 18789, binding to 127.0.0.1
- AI Model / Channel / API_KEY / Skills / Hooks: Configure as needed or skip all for now, you can configure models and keys uniformly in "Connect to OriginRouter" later
Startup Methods
Two optional methods:
| Method | Use Case | Operation |
|---|---|---|
| Foreground | Quick testing, don't want to use admin | Run openclaw gateway in terminal, don't close this window, then access console via browser |
| Daemon | Long-term use, auto-start on boot | Open PowerShell as administrator, re-run openclaw onboard or openclaw gateway install, then openclaw gateway start |
Control UI Information (displayed at the end of the wizard, please save it):
- Web UI:
http://127.0.0.1:18789/ - Web UI (with token):
http://127.0.0.1:18789/#token=your_token - Gateway WS:
ws://127.0.0.1:18789
Please always use the link with token to open the console, otherwise you may encounter "token_missing" or "too many failed authentication attempts".
You can also run openclaw dashboard at any time to open the browser.
Edit Configuration File
Find the local configuration file openclaw.json:
- Windows:
C:\Users\YourUsername\.openclaw\openclaw.json - macOS / Linux:
~/.openclaw/openclaw.json
Depending on your subscription plan, copy the matching template below verbatim:
openclaw.json
For users on the pay-as-you-go API plan. Requests are served via the /beta/v1 endpoint and billed per request.
{
"models": {
"mode": "merge",
"providers": {
"originrouter": {
"baseUrl": "https://api.easytransnote.com/beta/v1",
"apiKey": "${MY_API_KEY}",
"api": "openai-completions",
"models": [
{
"id": "gpt-5.4",
"name": "GPT-5.4",
"reasoning": true,
"input": ["text", "image"],
"cost": { "input": 2.5, "output": 15, "cacheRead": 0.25, "cacheWrite": 0.25 },
"contextWindow": 400000,
"maxTokens": 128000
},
{
"id": "gemini-3.1-pro-preview",
"name": "Gemini 3.1 Pro Preview",
"reasoning": true,
"input": ["text", "image"],
"cost": { "input": 2, "output": 12, "cacheRead": 0.2, "cacheWrite": 0.2 },
"contextWindow": 1048576,
"maxTokens": 65536
}
]
}
}
}
}
With an OriginRouter One subscription, choose the corresponding Coding API endpoint and make sure the model ID comes from the Supported Models list; with the pay-as-you-go API plan, choose the corresponding Beta API endpoint and make sure the model ID comes from the Model List; a plan and endpoint mismatch may make models unavailable or result in unexpected billing.
Complete Setup
OpenClaw provides more channels for you to interact with the bot, such as Web UI, Discord, Slack, etc. You can set up these channels by referring to the official documentation: Channels Setup
- For Web UI, you can access it by opening the Web UI (with token) link displayed in the terminal
Post-Installation Verification
Verify everything is working correctly:
openclaw doctor # Check configuration issues
openclaw status # View gateway status
openclaw dashboard # Open Dashboard in browser