Create response

POST/openai/v1/responses

Responses-style body (input, instructions, tools, stream) mapped to the same pipeline as chat completions. Same auth, tokensaver object, and X-Tokensaver-* headers.

Base URL prefix: https://api.tokensaver.fr/openai/v1. Non-stream returns an object with object: "response"; stream uses SSE with response.* events.

Module defaults and extensions match Chat completions and Headers.

curl -sS "https://api.tokensaver.fr/openai/v1/responses" \
  -H "Authorization: Bearer $TS_KEY" -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-4o","input":"Hello","stream":false}'