Start with the workflow you use
Codex CLI
Configure a custom model provider with an environment-based key and the Responses wire API.
Cursor
Test standard chat scenarios using the custom API settings exposed by your Cursor version.
API clients
Use curl, Python, Node.js, or another client that accepts a custom Base URL.
Operational visibility without hidden claims
/v1/modelsDiscover model IDs currently available to your account before building a request.
/v1/responsesSend standard or streamed Responses-compatible requests where the selected model supports them.
DashboardInspect usage records, remaining platform credits, and account-level information.
Service relationship and compatibility
Quick AI Coding is not affiliated with, endorsed by, sponsored by, or an official partner of OpenAI or Cursor. “OpenAI-compatible” refers only to selected request and response formats tested by Quick AI Coding; it does not mean full feature parity with the OpenAI API.
Models, rates, limits, and tested fields can change. Check /v1/models and the current dashboard before production use. Regional eligibility and the service terms apply.