Model IDs and options in examples are illustrative. Select an available model from GET /v1/models and check its capabilities and current prices.

POST /v1/responses

OpenAI Responses API format. Requests are translated internally to Chat Completions and back.

Use text.format for JSON output. A json_schema format carries name, schema and optional strict directly. When response healing is enabled, non-streaming JSON responses use the same repair and status headers as Chat Completions.

Parameter	Type	Default	Description
model	string	required	Model ID
input	string \| array	required	User input or conversation array
instructions	string	—	System instructions
stream	boolean	false	Enable SSE streaming
reasoning	object	—	{effort: "low"\|"medium"\|"high", summary: "auto"}

Example

curl -X POST https://relayon.app/v1/responses \
  -H "Authorization: Bearer ro_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "alims-intl/qwen-plus",
    "input": "Explain recursion in one sentence.",
    "instructions": "Be concise.",
    "reasoning": {"effort": "medium"}
  }'