Function Calling - SaveGate Documentation

Overview

Function calling allows models to call external functions/tools, enabling them to interact with APIs, databases, and other systems.

Basic Function Calling

from openai import OpenAI
import json

client = OpenAI(
    api_key="sk-savegate-xxxxxxxxxxxxx",
    base_url="https://api.savegate.ai/v1"
)

# Define functions
functions = [
    {
        "name": "get_weather",
        "description": "Get the current weather in a location",
        "parameters": {
            "type": "object",
            "properties": {
                "location": {
                    "type": "string",
                    "description": "City and state, e.g. San Francisco, CA"
                },
                "unit": {
                    "type": "string",
                    "enum": ["celsius", "fahrenheit"]
                }
            },
            "required": ["location"]
        }
    }
]

# Make request
response = client.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": "What's the weather in Boston?"}],
    functions=functions,
    function_call="auto"
)

# Check if function was called
if response.choices[0].message.function_call:
    function_call = response.choices[0].message.function_call
    print(f"Function: {function_call.name}")
    print(f"Arguments: {function_call.arguments}")

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'sk-savegate-xxxxxxxxxxxxx',
  baseURL: 'https://api.savegate.ai/v1'
});

const functions = [
  {
    name: 'get_weather',
    description: 'Get the current weather in a location',
    parameters: {
      type: 'object',
      properties: {
        location: {
          type: 'string',
          description: 'City and state, e.g. San Francisco, CA'
        },
        unit: {
          type: 'string',
          enum: ['celsius', 'fahrenheit']
        }
      },
      required: ['location']
    }
  }
];

const response = await client.chat.completions.create({
  model: 'gpt-4',
  messages: [{ role: 'user', content: "What's the weather in Boston?" }],
  functions: functions,
  function_call: 'auto'
});

if (response.choices[0].message.function_call) {
  const functionCall = response.choices[0].message.function_call;
  console.log(`Function: ${functionCall.name}`);
  console.log(`Arguments: ${functionCall.arguments}`);
}

Complete Example with Function Execution

import json
import requests

def get_weather(location, unit="fahrenheit"):
    """Actual implementation of weather lookup"""
    # This would call a real weather API
    # For demo, return mock data
    return {
        "location": location,
        "temperature": 72,
        "unit": unit,
        "forecast": "Sunny"
    }

def run_conversation(user_message):
    # Step 1: Send message with function definitions
    messages = [{"role": "user", "content": user_message}]

    response = client.chat.completions.create(
        model="gpt-4",
        messages=messages,
        functions=[{
            "name": "get_weather",
            "description": "Get current weather",
            "parameters": {
                "type": "object",
                "properties": {
                    "location": {"type": "string"},
                    "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}
                },
                "required": ["location"]
            }
        }],
        function_call="auto"
    )

    message = response.choices[0].message

    # Step 2: Check if model wants to call a function
    if message.function_call:
        function_name = message.function_call.name
        arguments = json.loads(message.function_call.arguments)

        # Step 3: Call the function
        if function_name == "get_weather":
            function_response = get_weather(**arguments)

            # Step 4: Send function response back to model
            messages.append(message)
            messages.append({
                "role": "function",
                "name": function_name,
                "content": json.dumps(function_response)
            })

            # Step 5: Get final response
            second_response = client.chat.completions.create(
                model="gpt-4",
                messages=messages
            )

            return second_response.choices[0].message.content

    return message.content

# Usage
result = run_conversation("What's the weather like in Boston?")
print(result)
# Output: "The current weather in Boston is sunny with a temperature of 72°F."

Multiple Functions

functions = [
    {
        "name": "get_weather",
        "description": "Get weather for a location",
        "parameters": {
            "type": "object",
            "properties": {
                "location": {"type": "string"}
            },
            "required": ["location"]
        }
    },
    {
        "name": "search_web",
        "description": "Search the web for information",
        "parameters": {
            "type": "object",
            "properties": {
                "query": {"type": "string"}
            },
            "required": ["query"]
        }
    },
    {
        "name": "calculate",
        "description": "Perform mathematical calculations",
        "parameters": {
            "type": "object",
            "properties": {
                "expression": {"type": "string"}
            },
            "required": ["expression"]
        }
    }
]

# The model will choose the appropriate function
response = client.chat.completions.create(
    model="gpt-4",
    messages=[{"role": "user", "content": "What is 25 * 17?"}],
    functions=functions
)

Advanced: Function Calling Agent

class FunctionAgent:
    def __init__(self, client):
        self.client = client
        self.functions = {}
        self.function_schemas = []

    def register_function(self, schema, implementation):
        """Register a function with its schema and implementation"""
        self.function_schemas.append(schema)
        self.functions[schema["name"]] = implementation

    def execute_function(self, name, arguments):
        """Execute a registered function"""
        if name in self.functions:
            return self.functions[name](**arguments)
        raise ValueError(f"Function {name} not found")

    def run(self, user_message, max_iterations=5):
        """Run the agent with function calling support"""
        messages = [{"role": "user", "content": user_message}]

        for i in range(max_iterations):
            response = self.client.chat.completions.create(
                model="gpt-4",
                messages=messages,
                functions=self.function_schemas,
                function_call="auto"
            )

            message = response.choices[0].message

            # If no function call, we're done
            if not message.function_call:
                return message.content

            # Execute the function
            function_name = message.function_call.name
            arguments = json.loads(message.function_call.arguments)

            print(f"Calling {function_name} with {arguments}")

            result = self.execute_function(function_name, arguments)

            # Add to conversation
            messages.append(message)
            messages.append({
                "role": "function",
                "name": function_name,
                "content": json.dumps(result)
            })

        return "Max iterations reached"

# Usage
agent = FunctionAgent(client)

# Register functions
agent.register_function(
    {
        "name": "get_weather",
        "description": "Get weather",
        "parameters": {
            "type": "object",
            "properties": {
                "location": {"type": "string"}
            },
            "required": ["location"]
        }
    },
    get_weather
)

agent.register_function(
    {
        "name": "calculate",
        "description": "Calculate expression",
        "parameters": {
            "type": "object",
            "properties": {
                "expression": {"type": "string"}
            },
            "required": ["expression"]
        }
    },
    lambda expression: eval(expression)  # Be careful with eval!
)

# Run
result = agent.run("What's 25 + 17 and what's the weather in Boston?")
print(result)

More Examples

Learn how to use multiple models together