function calling api

If you mean function calling with an AI API, here’s the basic idea:

What it is

You define a tool/function in your app, send its schema to the model, and the model can respond with a request to call that function instead of plain text.

Typical flow

1. You define functions the model may use.
2. Send user message + function definitions to the model.
3. The model replies with either:
   • normal text, or
   • a structured function call like get_weather({ "city": "Paris" })
4. Your app executes the function
5. You send the function result back to the model
6. The model produces the final answer

---

Example concept

Suppose you define:

{
  "name": "get_weather",
  "description": "Get the current weather for a city",
  "parameters": {
    "type": "object",
    "properties": {
      "city": { "type": "string" }
    },
    "required": ["city"]
  }
}

User asks:

What’s the weather in London?

The model may return:

{
  "name": "get_weather",
  "arguments": {
    "city": "London"
  }
}

Your code calls your weather API, then passes the result back to the model.

---

If you want code

Tell me which API you’re using:

• OpenAI
• Anthropic
• Google Gemini
• Azure OpenAI
• another

and I can give you a working example in Python or JavaScript.