Skip to main content

​
Responses Function Calling

The Responses API supports Function Calling, allowing models to call functions you define to retrieve external data or perform actions. 
POST /v1/responses

​
Define Tools and Call

curl https://crazyrouter.com/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "gpt-4o",
    "input": "What is the weather like in Beijing and Shanghai today?",
    "tools": [
      {
        "type": "function",
        "name": "get_weather",
        "description": "Get the current weather for a specified city",
        "parameters": {
          "type": "object",
          "properties": {
            "city": {
              "type": "string",
              "description": "City name"
            }
          },
          "required": ["city"]
        }
      }
    ]
  }'

​
Function Call Response

When the model decides to call a function, the response contains function_call type outputs: 
{
  "id": "resp_abc123",
  "object": "response",
  "output": [
    {
      "type": "function_call",
      "id": "fc_001",
      "call_id": "call_abc123",
      "name": "get_weather",
      "arguments": "{\"city\": \"Beijing\"}"
    },
    {
      "type": "function_call",
      "id": "fc_002",
      "call_id": "call_def456",
      "name": "get_weather",
      "arguments": "{\"city\": \"Shanghai\"}"
    }
  ],
  "status": "incomplete"
}

​
Return Function Results

Return function execution results via function_call_output: 
{
  "model": "gpt-4o",
  "input": [
    {"role": "user", "content": "What is the weather like in Beijing and Shanghai today?"},
    {
      "type": "function_call",
      "call_id": "call_abc123",
      "name": "get_weather",
      "arguments": "{\"city\": \"Beijing\"}"
    },
    {
      "type": "function_call_output",
      "call_id": "call_abc123",
      "output": "{\"temperature\": 22, \"condition\": \"sunny\"}"
    },
    {
      "type": "function_call",
      "call_id": "call_def456",
      "name": "get_weather",
      "arguments": "{\"city\": \"Shanghai\"}"
    },
    {
      "type": "function_call_output",
      "call_id": "call_def456",
      "output": "{\"temperature\": 26, \"condition\": \"cloudy\"}"
    }
  ],
  "tools": [...]
}

​
Streaming Function Calling

Python
stream = client.responses.create(
    model="gpt-4o",
    input="Look up Apple's stock price",
    tools=tools,
    stream=True
)

for event in stream:
    if event.type == "response.function_call_arguments.delta":
        print(event.delta, end="")
    elif event.type == "response.function_call_arguments.done":
        print("\nFunction call complete:", event.arguments)
    elif event.type == "response.output_text.delta":
        print(event.delta, end="")
The main difference between Function Calling in the Responses API and the Chat Completions API is that tool definitions use name and parameters fields directly, without needing to nest them inside a function object.