User Guide
Keys, authentication, limits, errors, and billing
Once the Quick Start is working, these pages cover how the API behaves day to day: how keys work, which models you can call, what happens when you hit a limit, and how usage is billed.
Keys & authentication
Create and revoke keys, send Authorization, and read 401 / 403.
Models & parameters
List models, pick an id, and the Chat Completions fields you will use most.
Rate limits & errors
Rate-limit headers, backoff for 429, and the rest of the error table.
Usage & billing
The usage object, streaming token counts, and where to see remaining balance.
Also see
Quick Start
Your first request, the OpenAI SDK swap, and streaming.
Integrations
Codex, Cursor, Continue, Aider, Open WebUI, CrewAI, and the OpenAI SDK.
API reference
Every endpoint and parameter, with request and response schemas.
Status
Live availability and incident history.
Support
Reach the team for access requests, quota changes, and billing.