- Step1: Sign up on CodingPlanX and obtain an API key from https://codingplanx.ai/user/apiKey.
- Step2: Replace your Base URL and API key in existing OpenAI-compatible code or SDKs with CodingPlanX endpoints.
- Step3: Choose a specific model or enable Auto-mode to let the gateway route to the best provider.
- Step4: Configure request parameters (model, max_tokens, temperature, streaming) as usual and call the /v1/chat/completions endpoint.
- Step5: Monitor usage, logs, and billing in the dashboard; adjust model selection or concurrency settings as needed.
- Step6: Integrate with developer tools (Cursor, Claude Code, OpenClaw) or use provided examples and SDKs for quick adoption.