Backend HTTP integration

Use a server process to send the documented JSON request, check its HTTP status and validate the response before using model output. Keep request timeouts and usage limits explicit.

Runnable integrations

The repository provides examples/chat.py for Python standard-library HTTP and examples/chat.mjs for Node native fetch. Both read secrets from environment variables and use controlled fixtures for success and failure tests. Follow the complete Python walkthrough.

Request shape

POST /v1/chat/completions accepts model and messages, with optional output and sampling controls. A non-streaming text example reads choices[0].message.content; a tool-only response or malformed body needs a separate handling path.

Production boundary

Authenticate your own application callers, constrain model and output choices, and sanitize diagnostics. A timeout is an uncertain outcome. These are HTTP examples rather than a claim that a dedicated AIx SDK exists. See SDK status.

Verification and next steps

These instructions describe inspected repository contracts and safe integration boundaries. They do not prove live account access or external provider behavior. Continue with developer answers, API overview and verification limits.