Responses

POST /v1/responses

Use the model name shown on the Pricing page.

Authentication

Authorization: Bearer OMROUTER_API_KEY
Content-Type: application/json

Request body

Field	Type	Required	Description
model	string	Yes	Model name shown on the Pricing page.
input	string or array	Yes	Text or structured input items.
instructions	string or object	No	Overall instructions for the model.
max_output_tokens	integer	No	Maximum output tokens where supported.
stream	boolean	No	Stream response events when true.
tools	array	No	Tools supported by the selected model and route.
tool_choice	string or object	No	Tool selection control where supported.
reasoning	object	No	Reasoning configuration where supported.
previous_response_id	string	No	Continue a compatible previous response.

Example

curl --silent --show-error \
  'https://omrouter.com/v1/responses' \
  --header "Authorization: Bearer $OMROUTER_API_KEY" \
  --header 'Content-Type: application/json' \
  --data "{
    \"model\": \"$OMROUTER_MODEL_ID\",
    \"input\": \"Reply with one short sentence.\"
  }"

Response

Generated content is in the output array. OpenAI SDKs usually also provide a direct way to read the output text.

{
  "id": "resp_...",
  "object": "response",
  "status": "completed",
  "model": "model-id",
  "output": [
    {
      "type": "message",
      "role": "assistant",
      "content": [
        {"type": "output_text", "text": "..."}
      ]
    }
  ],
  "usage": {}
}

The content of output changes with the request type and model.

Compact responses

POST /v1/responses/compact is for clients that explicitly support Compact Responses. Normal Responses clients should continue to use /v1/responses.

Optional features

Tools, reasoning, prompt caching, storage, and continuation depend on the selected model. If the API reports an unsupported parameter, remove that optional field and try again.