Skip to content
Search documentation titles, headings, and guides
API v1Updated 2026-09-16

Quickstart

Send your first request in under five minutes: sign in, create an API key, call a model, and track your usage.

This guide takes you from a fresh account to a working inference request. No credit card is required to get started — you only need a Google account and a browser.

1. Sign in with Google#

Open the dashboard and click Sign in with Google. You authenticate with an existing Google account, so there is no separate zrouter password to manage. After signing in you land in your workspace.

2. Create an API key#

From your workspace, open the API keys section and click Create API key. The key looks like zr_<public-id>_<secret> and gives your code access to every inference endpoint.

You can revoke a key and issue a fresh one at any time from the same screen, with no disruption to your other keys.

3. Make your first request#

Send a chat completion to the gateway's OpenAI-compatible endpoint. The examples below use the official OpenAI SDK, but any OpenAI- or Anthropic-compatible client works — you only change the base URL.

curl -X POST "https://api.zrouter.dev/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<public-name>",
    "messages": [{"role": "user", "content": "Say hello!"}]
  }'

4. Track usage in your workspace#

Every request is metered and settled immediately. Head back to your workspace to see your current Token Plan and PAYG balances, review recent usage, and export your transactions or balances as CSV or JSON for reconciliation.

5. Next steps#