OpenAI Function Calling Explained (Simple Guide)

What function calling is, how the schema-driven tool flow works, and a minimal weather example end-to-end.

Aug 11, 2024 · 4 min read · 1616 Words · -Views -Comments · AI Tutorials
AI
Read with AI

Choose an AI assistant

ChatGPT Open in a new tab with the article link and prompt
Claude Copy the prompt, then open Claude

Open Claude
Trouble reading the article?

If the prompt is missing, copy the full prompt below into your AI assistant. If the AI cannot read the article, paste the visible passage you want to discuss and add your question.

Many “agent” products have launched in China — Tencent’s 元宝/元器, ByteDance’s 豆包, Moonshot’s Kimi+, etc. A key technology behind them is function calling. So what is function calling?

这里我实际使用了下混元和OpenAI的,来梳理下function call怎么玩的。

https://cdn.openai.com/API/docs/images/function-calling-diagram.png

Function Call

先说概念,顾名思义,即函数调用。

Note: from the API’s perspective, the model doesn’t directly run the function. It selects a function and arguments based on your predefined schema and the user’s input.

So function calling is a form of intent recognition.

Interaction flow

  1. 定义工具列表

          tools: [{
            type: "function", function: {
              name: "get_realtime_weather", summary: "获取当前的天气情况", parameters: {
                type: "object", properties: {
                  city: {
                    type: "string", summary: "城市名称"
                  }
                }, required: ["city"], additionalProperties: true,
              }
            }
          }], // 工具选项,当为string类型时,可选参数为"none"/"auto", "none"表示不会使用工具,"auto"表示模型自行选择生成回复或调用工具;当为object类型时,强制模型调用指定的function
          tool_choice: "auto"
    

In this example, the function requires only a city parameter — and it’s marked as required.

  1. The user asks “What should I wear tomorrow?”. The model infers it’s weather‑related and, since the city is required, prompts for it.

    为了给您提供合适的穿衣建议,我需要知道您所在的城市。请问您能告诉我您所在的城市名称吗?(城市名称为必填项)

  2. User replies “北京 (Beijing)”.

  3. The model emits a tool call for get_realtime_weather with arguments.

    {
      id: 'call_cqs8e7k2c3m5dtj9a9tg',
      type: 'function',
      function: { name: 'get_realtime_weather', arguments: '{"city":"北京"}' }
    }
    
  4. Your code executes the function and returns the result back to the model.

    {
      "role": "tool",
      "content": "北京 天气晴朗,微风,最高气温27摄氏度,最低气温17摄氏度"
    }
    
  5. The model composes a final answer for the user based on the tool result.

    {"created":1723369252,"id":"b097e75efe3d328fa0f6ecc1f884e908","model":"hunyuan","version":"202403121316","choices":[{"finish_reason":"stop","message":{"role":"assistant","content":"根据查询到的天气情况,北京明天天气晴朗,微风,最高气温27摄氏度,最低气温17摄氏度。建议您可以穿着长袖衬衫或薄型外套,搭配长裤。如果您对温度比较敏感,可以随身携带一件轻薄的外01,"completion_tokens":50,"total_tokens":151}}
    

That’s a complete function‑calling round‑trip.

这里贴下完整的message信息

 [
  { role: 'user', content: '我明天应该穿什么?' },
  {
    role: 'assistant',
    content: '为了给您提供合适的穿衣建议,我需要知道您所在的城市。请问您能告诉我您所在的城市名称吗?(城市名称为必填项)'
  },
  { role: 'user', content: '北京' },
  {
    role: 'assistant',
    content: '调用天气查询工具,输入参数为用户所在的城市北京。\n' +
      '\t\n' +
      '\t现在已经获得了用户所在的城市信息,可以使用天气查询工具来获取明天的天气情况。根据用户提供的城市信息和当前日期,可以计算出明天的日期,并查询那天的天气情况。',
    tool_calls: [ [Object] ]
  },
  { role: 'tool', content: '北京 天气晴朗,微风,最高气温27摄氏度,最低气温17摄氏度' },
  {
    role: 'assistant',
    content: '根据查询到的天气情况,明天北京天气晴朗,微风,最高气温27摄氏度,最低气温17摄氏度。建议您可以穿着长袖衬衫或薄型外套,搭配长裤。如果您对温度比较敏感,可以随身携带一件轻薄的外套以防温度变化。'
  }
]
https://static.1991421.cn/2024/2024-08-11-180536.jpeg

Agent

AI platforms support visual agent builders that compose multiple capabilities; function calling is the key for intent recognition and tool invocation.

With function calling, you can integrate and call any tool/function you expose.

Final Thoughts

If you haven’t tried function calling yet, give it a spin.

Authors
Developer, digital product enthusiast, tinkerer, sharer, open source lover