Create Model Response

Use the POST /v1/responses endpoint to create model responses, supporting standard, streaming, and thinking control

Create Model Response

POST /v1/responses

The Responses API provides a more concise interface for creating model responses.

Authentication

Authorization: Bearer YOUR_API_KEY

---

Basic Request

cURL

curl https://crazyrouter.com/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gpt-4o",
    "input": "Explain what an API is in one sentence"
  }'

Response Format

{
  "id": "resp_abc123",
  "object": "response",
  "created_at": 1709123456,
  "model": "gpt-4o",
  "output": [
    {
      "type": "message",
      "id": "msg_abc123",
      "role": "assistant",
      "content": [
        {
          "type": "output_text",
          "text": "An API is a set of rules and protocols that define how software components interact, enabling different programs to communicate and share data."
        }
      ]
    }
  ],
  "usage": {
    "input_tokens": 15,
    "output_tokens": 40,
    "total_tokens": 55
  }
}

---

Multi-Turn Conversation

Use the input array to pass conversation history:

Python

response = client.responses.create(
    model="gpt-4o",
    input=[
        {"role": "user", "content": "I want to learn programming"},
        {"role": "assistant", "content": "Great! Which programming language would you like to learn?"},
        {"role": "user", "content": "Python, where should I start?"}
    ]
)

print(response.output_text)

---

Streaming Request

cURL

curl https://crazyrouter.com/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gpt-4o",
    "input": "Write a poem about programming",
    "stream": true
  }'

Streaming Event Types

Event Type	Description
response.created	Response created
response.in_progress	Response processing
response.output_item.added	New output item added
response.content_part.added	New content block added
response.output_text.delta	Text increment
response.output_text.done	Text complete
response.content_part.done	Content block complete
response.output_item.done	Output item complete
response.completed	Response complete

---

Thinking Control

Use the reasoning parameter to control reasoning depth:

Python

response = client.responses.create(
    model="o4-mini",
    input="Prove the Pythagorean theorem",
    reasoning={
        "effort": "high"
    }
)

print(response.output_text)

reasoning Parameter

Field	Type	Description
effort	string	Reasoning depth: low, medium, high
summary	string	Thinking summary mode: auto, concise, detailed

---

Request Parameters

Parameter	Type	Required	Description
model	string	Yes	Model name
input	string|array	Yes	Input content or message list
stream	boolean	No	Whether to stream output
temperature	number	No	Sampling temperature
max_output_tokens	integer	No	Maximum output tokens
tools	array	No	Available tools list
reasoning	object	No	Reasoning control
instructions	string	No	System instructions (similar to system message)

The Responses API’s input parameter can be either a simple string or a message array, providing more flexibility.