Use Your ChatGPT Plus Subscription with OpenClaw

Running Hermes Agent instead of OpenClaw? See the Hermes ChatGPT subscription guide.

How It Works

OpenClaw supports the openai provider, which routes requests through your ChatGPT subscription instead of the pay-per-token OpenAI API. You sign in with your OpenAI account via OAuth, and OpenClaw uses your subscription's included quota — no API billing.

What You Need

• An active ChatGPT Plus ($20/mo) or ChatGPT Pro ($200/mo) subscription
• A deployed OpenClaw instance on OpenClaw Launch

Step 1: Open the Terminal

1. Go to your Dashboard and find your instance.
2. Click the Terminal button to open the web terminal.

Step 2: Run the Login Command

In the terminal, run:

openclaw models auth login --provider openai --device-code

You'll see a URL and a short code. Open the URL in your local browser (on your computer, not inside the terminal).

Step 3: Sign In with Your OpenAI Account

1. The URL opens an OpenAI login page. Sign in with the account that has your ChatGPT Plus subscription.
2. Enter the code shown in your terminal when OpenAI asks for it, and approve the connection.

Prefer the classic browser-callback flow? Run the command without --device-code: after signing in, your browser redirects to a localhost URL and shows a "This site can't be reached" error — that's normal (the terminal runs on a remote server). Copy the full http://localhost:1455/auth/callback?code=... URL from the address bar and paste it back into the terminal.

Step 4: Set GPT-5.5 as Your Default Model

Run this command to switch your agent to GPT-5.5:

openclaw models set openai/gpt-5.5

That's it. Your OpenClaw agent now uses GPT-5.5 through your ChatGPT subscription.

Step 5: Verify It Works

Send a message to your bot on Telegram, Discord, or whichever platform you connected. The response should come from GPT-5.5. You can also check the active model with:

openclaw models status

Available Models

Important Notes

• Weekly quota limits: ChatGPT Plus has a weekly usage cap for Codex access. Heavy usage may hit limits before the week resets.
• Token refresh: OpenAI uses rotating refresh tokens. Occasionally you may need to re-run the login command if the session expires.
• For production workloads: If you need guaranteed uptime and no quota limits, consider using an OpenAI API key instead.

Other Ways to Bring Your Own Key

Don't have ChatGPT Plus? You can still bring your own API key from any provider:

• OpenAI API — Pay-per-token, no subscription needed
• Anthropic Claude — Claude Sonnet 4.6, Opus 4.6
• Google Gemini — Gemini 3.1 Pro, Flash
• OpenRouter — Access 200+ models with one key

Or skip API keys entirely — OpenClaw Launch includes AI credits and free models out of the box.

Troubleshooting

"This site can't be reached" after signing in

Only happens on the browser-callback flow (without --device-code). Your browser redirects to localhost, but the terminal runs on a remote server. Simply copy the full URL from your browser's address bar and paste it into the terminal — it will extract the authorization code automatically. Or re-run with --device-code to avoid this entirely.

Token refresh errors

If you see refresh_token_reused errors, re-run the login command:

openclaw models auth login --provider openai --device-code

Quota exceeded

ChatGPT Plus has weekly limits. If you hit the cap, your bot will return errors until the quota resets. Upgrade to ChatGPT Pro ($200/mo) for higher limits, or switch to an API key for unlimited usage.