Step 1: Sign in and open the console

After signing in, open the console from the top navigation. Keys, balance, and usage records are all managed there.

Step 2: Create an API key

  1. Open the API Keys page.
  2. Click Add key and give it a recognizable note (for example: local development / production).
  3. The full key is shown only once after creation. Save it to a server-side environment variable or a password manager immediately.
  4. Never commit keys to a repository, share them in screenshots, or hard-code them in browser code.

Step 3: Choose a currently available model

Open the Live pricing section on the home page: only models marked Live settlement are callable. Rows marked as reference are for comparing official prices only and cannot be called. You can also list models directly:

curl https://47.237.177.122/v1/models \
  -H "Authorization: Bearer sk-YOUR_KEY"

Step 4: Send your first request

Copy the returned model ID and run a minimal Chat Completions request:

curl https://47.237.177.122/v1/chat/completions \
  -H "Authorization: Bearer sk-YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "THE_MODEL_ID_YOU_COPIED",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

A 200 response with a choices field means the whole chain from gateway to upstream works.

Step 5: Wire up your clients

  • Claude Code: see the Client Setup guide.
  • Codex CLI: see the Client Setup guide.
  • Other OpenAI-compatible apps: set baseURL to https://47.237.177.122/v1 and apiKey to your key.

Quick error lookup

  • 401: invalid or expired key. Re-copy it and check the request header.
  • 403: insufficient balance or permission. Top up or check quota and group access.
  • 429: rate limited. Retry with exponential backoff, see the Troubleshooting guide.
  • Model unavailable: confirm the ID comes from /v1/models and is marked Live settlement, then read the error body for details.