> ## Documentation Index
> Fetch the complete documentation index at: https://docs.apimart.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenAI 多模态响应接口

>  - 完全兼容 OpenAI Responses API 格式
- 支持文本和图像的多模态输入
- 支持工具扩展：网络搜索、文件搜索、函数调用、远程MCP 

<RequestExample>
  ```bash cURL theme={null}
  curl https://api.apimart.ai/v1/responses \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer <token>" \
    -d '{
      "model": "gpt-5.2-pro",
      "input": [
        {
          "role": "user",
          "content": [
            {
              "type": "input_text",
              "text": "这张图片里有什么？"
            },
            {
              "type": "input_image",
              "image_url": "https://openai-documentation.vercel.app/images/cat_and_otter.png"
            }
          ]
        }
      ]
    }'
  ```

  ```python Python theme={null}
  import requests
  import os

  url = "https://api.apimart.ai/v1/responses"

  payload = {
      "model": "gpt-5.2-pro",
      "input": [
          {
              "role": "user",
              "content": [
                  {
                      "type": "input_text",
                      "text": "这张图片里有什么？"
                  },
                  {
                      "type": "input_image",
                      "image_url": "https://openai-documentation.vercel.app/images/cat_and_otter.png"
                  }
              ]
          }
      ]
  }

  headers = {
      "Authorization": f"Bearer {os.environ.get('OPENAI_API_KEY')}",
      "Content-Type": "application/json"
  }

  response = requests.post(url, json=payload, headers=headers)

  print(response.json())
  ```

  ```javascript JavaScript theme={null}
  const url = "https://api.apimart.ai/v1/responses";

  const payload = {
    model: "gpt-5.2-pro",
    input: [
      {
        role: "user",
        content: [
          {
            type: "input_text",
            text: "这张图片里有什么？"
          },
          {
            type: "input_image",
            image_url: "https://openai-documentation.vercel.app/images/cat_and_otter.png"
          }
        ]
      }
    ]
  };

  const headers = {
    "Authorization": `Bearer ${process.env.OPENAI_API_KEY}`,
    "Content-Type": "application/json"
  };

  fetch(url, {
    method: "POST",
    headers: headers,
    body: JSON.stringify(payload)
  })
    .then(response => response.json())
    .then(data => console.log(data))
    .catch(error => console.error('Error:', error));
  ```

  ```go Go theme={null}
  package main

  import (
      "bytes"
      "encoding/json"
      "fmt"
      "io/ioutil"
      "net/http"
      "os"
  )

  func main() {
      url := "https://api.apimart.ai/v1/responses"

      payload := map[string]interface{}{
          "model": "gpt-5.2-pro",
          "input": []map[string]interface{}{
              {
                  "role": "user",
                  "content": []map[string]string{
                      {
                          "type": "input_text",
                          "text": "这张图片里有什么？",
                      },
                      {
                          "type":      "input_image",
                          "image_url": "https://openai-documentation.vercel.app/images/cat_and_otter.png",
                      },
                  },
              },
          },
      }

      jsonData, _ := json.Marshal(payload)

      req, _ := http.NewRequest("POST", url, bytes.NewBuffer(jsonData))
      req.Header.Set("Authorization", "Bearer "+os.Getenv("OPENAI_API_KEY"))
      req.Header.Set("Content-Type", "application/json")

      client := &http.Client{}
      resp, err := client.Do(req)
      if err != nil {
          panic(err)
      }
      defer resp.Body.Close()

      body, _ := ioutil.ReadAll(resp.Body)
      fmt.Println(string(body))
  }
  ```

  ```java Java theme={null}
  import java.net.http.HttpClient;
  import java.net.http.HttpRequest;
  import java.net.http.HttpResponse;
  import java.net.URI;

  public class Main {
      public static void main(String[] args) throws Exception {
          String url = "https://api.apimart.ai/v1/responses";
          String apiKey = System.getenv("OPENAI_API_KEY");

          String payload = """
          {
            "model": "gpt-5.2-pro",
            "input": [
              {
                "role": "user",
                "content": [
                  {
                    "type": "input_text",
                    "text": "这张图片里有什么？"
                  },
                  {
                    "type": "input_image",
                    "image_url": "https://openai-documentation.vercel.app/images/cat_and_otter.png"
                  }
                ]
              }
            ]
          }
          """;

          HttpClient client = HttpClient.newHttpClient();
          HttpRequest request = HttpRequest.newBuilder()
              .uri(URI.create(url))
              .header("Authorization", "Bearer " + apiKey)
              .header("Content-Type", "application/json")
              .POST(HttpRequest.BodyPublishers.ofString(payload))
              .build();

          HttpResponse<String> response = client.send(request,
              HttpResponse.BodyHandlers.ofString());

          System.out.println(response.body());
      }
  }
  ```

  ```php PHP theme={null}
  <?php

  $url = "https://api.apimart.ai/v1/responses";
  $apiKey = getenv('OPENAI_API_KEY');

  $payload = [
      "model" => "gpt-5.2-pro",
      "input" => [
          [
              "role" => "user",
              "content" => [
                  [
                      "type" => "input_text",
                      "text" => "这张图片里有什么？"
                  ],
                  [
                      "type" => "input_image",
                      "image_url" => "https://openai-documentation.vercel.app/images/cat_and_otter.png"
                  ]
              ]
          ]
      ]
  ];

  $ch = curl_init($url);
  curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  curl_setopt($ch, CURLOPT_POST, true);
  curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($payload));
  curl_setopt($ch, CURLOPT_HTTPHEADER, [
      "Authorization: Bearer " . $apiKey,
      "Content-Type: application/json"
  ]);

  $response = curl_exec($ch);
  curl_close($ch);

  echo $response;
  ?>
  ```

  ```ruby Ruby theme={null}
  require 'net/http'
  require 'json'
  require 'uri'

  url = URI("https://api.apimart.ai/v1/responses")
  api_key = ENV['OPENAI_API_KEY']

  payload = {
    model: "gpt-5.2-pro",
    input: [
      {
        role: "user",
        content: [
          {
            type: "input_text",
            text: "这张图片里有什么？"
          },
          {
            type: "input_image",
            image_url: "https://openai-documentation.vercel.app/images/cat_and_otter.png"
          }
        ]
      }
    ]
  }

  http = Net::HTTP.new(url.host, url.port)
  http.use_ssl = true

  request = Net::HTTP::Post.new(url)
  request["Authorization"] = "Bearer #{api_key}"
  request["Content-Type"] = "application/json"
  request.body = payload.to_json

  response = http.request(request)
  puts response.body
  ```

  ```swift Swift theme={null}
  import Foundation

  let url = URL(string: "https://api.apimart.ai/v1/responses")!
  let apiKey = ProcessInfo.processInfo.environment["OPENAI_API_KEY"] ?? ""

  let payload: [String: Any] = [
      "model": "gpt-5.2-pro",
      "input": [
          [
              "role": "user",
              "content": [
                  [
                      "type": "input_text",
                      "text": "这张图片里有什么？"
                  ],
                  [
                      "type": "input_image",
                      "image_url": "https://openai-documentation.vercel.app/images/cat_and_otter.png"
                  ]
              ]
          ]
      ]
  ]

  var request = URLRequest(url: url)
  request.httpMethod = "POST"
  request.setValue("Bearer \(apiKey)", forHTTPHeaderField: "Authorization")
  request.setValue("application/json", forHTTPHeaderField: "Content-Type")
  request.httpBody = try? JSONSerialization.data(withJSONObject: payload)

  let task = URLSession.shared.dataTask(with: request) { data, response, error in
      if let error = error {
          print("Error: \(error)")
          return
      }
      
      if let data = data, let responseString = String(data: data, encoding: .utf8) {
          print(responseString)
      }
  }

  task.resume()
  ```

  ```csharp C# theme={null}
  using System;
  using System.Net.Http;
  using System.Text;
  using System.Threading.Tasks;

  class Program
  {
      static async Task Main(string[] args)
      {
          var url = "https://api.apimart.ai/v1/responses";
          var apiKey = Environment.GetEnvironmentVariable("OPENAI_API_KEY");

          var payload = @"{
              ""model"": ""gpt-5.2-pro"",
              ""input"": [
                  {
                      ""role"": ""user"",
                      ""content"": [
                          {
                              ""type"": ""input_text"",
                              ""text"": ""这张图片里有什么？""
                          },
                          {
                              ""type"": ""input_image"",
                              ""image_url"": ""https://openai-documentation.vercel.app/images/cat_and_otter.png""
                          }
                      ]
                  }
              ]
          }";

          using var client = new HttpClient();
          client.DefaultRequestHeaders.Add("Authorization", $"Bearer {apiKey}");

          var content = new StringContent(payload, Encoding.UTF8, "application/json");
          var response = await client.PostAsync(url, content);
          var result = await response.Content.ReadAsStringAsync();

          Console.WriteLine(result);
      }
  }
  ```

  ```c C theme={null}
  #include <stdio.h>
  #include <curl/curl.h>
  #include <stdlib.h>

  int main(void) {
      CURL *curl;
      CURLcode res;
      const char *api_key = getenv("OPENAI_API_KEY");

      curl_global_init(CURL_GLOBAL_DEFAULT);
      curl = curl_easy_init();

      if(curl) {
          const char *url = "https://api.apimart.ai/v1/responses";
          const char *payload = "{"
              "\"model\":\"gpt-5.2-pro\","
              "\"input\":[{\"role\":\"user\",\"content\":[{\"type\":\"input_text\",\"text\":\"这张图片里有什么？\"},{\"type\":\"input_image\",\"image_url\":\"https://openai-documentation.vercel.app/images/cat_and_otter.png\"}]}]"
          "}";

          char auth_header[256];
          snprintf(auth_header, sizeof(auth_header), "Authorization: Bearer %s", api_key);

          struct curl_slist *headers = NULL;
          headers = curl_slist_append(headers, auth_header);
          headers = curl_slist_append(headers, "Content-Type: application/json");

          curl_easy_setopt(curl, CURLOPT_URL, url);
          curl_easy_setopt(curl, CURLOPT_POSTFIELDS, payload);
          curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);

          res = curl_easy_perform(curl);

          if(res != CURLE_OK) {
              fprintf(stderr, "curl_easy_perform() failed: %s\n",
                      curl_easy_strerror(res));
          }

          curl_slist_free_all(headers);
          curl_easy_cleanup(curl);
      }

      curl_global_cleanup();
      return 0;
  }
  ```

  ```objectivec Objective-C theme={null}
  #import <Foundation/Foundation.h>

  int main(int argc, const char * argv[]) {
      @autoreleasepool {
          NSURL *url = [NSURL URLWithString:@"https://api.apimart.ai/v1/responses"];
          NSString *apiKey = [NSProcessInfo processInfo].environment[@"OPENAI_API_KEY"];
          
          NSDictionary *payload = @{
              @"model": @"gpt-5.2-pro",
              @"input": @[
                  @{
                      @"role": @"user",
                      @"content": @[
                          @{
                              @"type": @"input_text",
                              @"text": @"这张图片里有什么？"
                          },
                          @{
                              @"type": @"input_image",
                              @"image_url": @"https://openai-documentation.vercel.app/images/cat_and_otter.png"
                          }
                      ]
                  }
              ]
          };
          
          NSError *error;
          NSData *jsonData = [NSJSONSerialization dataWithJSONObject:payload
                                                            options:0
                                                              error:&error];
          
          NSMutableURLRequest *request = [NSMutableURLRequest requestWithURL:url];
          [request setHTTPMethod:@"POST"];
          [request setValue:[NSString stringWithFormat:@"Bearer %@", apiKey] 
              forHTTPHeaderField:@"Authorization"];
          [request setValue:@"application/json" forHTTPHeaderField:@"Content-Type"];
          [request setHTTPBody:jsonData];
          
          NSURLSessionDataTask *task = [[NSURLSession sharedSession] 
              dataTaskWithRequest:request
              completionHandler:^(NSData *data, NSURLResponse *response, NSError *error) {
                  if (error) {
                      NSLog(@"Error: %@", error);
                      return;
                  }
                  NSString *result = [[NSString alloc] initWithData:data 
                                                          encoding:NSUTF8StringEncoding];
                  NSLog(@"%@", result);
              }];
          
          [task resume];
          [[NSRunLoop mainRunLoop] run];
      }
      return 0;
  }
  ```

  ```ocaml OCaml theme={null}
  (* Requires cohttp and yojson libraries *)
  open Lwt
  open Cohttp
  open Cohttp_lwt_unix

  let url = "https://api.apimart.ai/v1/responses"
  let api_key = Sys.getenv "OPENAI_API_KEY"

  let payload = {|{
    "model": "gpt-5.2-pro",
    "input": [
      {
        "role": "user",
        "content": [
          {
            "type": "input_text",
            "text": "这张图片里有什么？"
          },
          {
            "type": "input_image",
            "image_url": "https://openai-documentation.vercel.app/images/cat_and_otter.png"
          }
        ]
      }
    ]
  }|}

  let () =
    let headers = Header.init ()
      |> fun h -> Header.add h "Authorization" ("Bearer " ^ api_key)
      |> fun h -> Header.add h "Content-Type" "application/json"
    in
    let body = Cohttp_lwt.Body.of_string payload in
    
    let response = Client.post ~headers ~body (Uri.of_string url) >>= fun (resp, body) ->
      body |> Cohttp_lwt.Body.to_string >|= fun body_str ->
      print_endline body_str
    in
    Lwt_main.run response
  ```

  ```dart Dart theme={null}
  import 'dart:convert';
  import 'dart:io';
  import 'package:http/http.dart' as http;

  void main() async {
    final url = Uri.parse('https://api.apimart.ai/v1/responses');
    final apiKey = Platform.environment['OPENAI_API_KEY'];
    
    final payload = {
      'model': 'gpt-5.2-pro',
      'input': [
        {
          'role': 'user',
          'content': [
            {
              'type': 'input_text',
              'text': '这张图片里有什么？'
            },
            {
              'type': 'input_image',
              'image_url': 'https://openai-documentation.vercel.app/images/cat_and_otter.png'
            }
          ]
        }
      ]
    };
    
    final response = await http.post(
      url,
      headers: {
        'Authorization': 'Bearer $apiKey',
        'Content-Type': 'application/json',
      },
      body: jsonEncode(payload),
    );
    
    print(response.body);
  }
  ```

  ```r R theme={null}
  library(httr)
  library(jsonlite)

  url <- "https://api.apimart.ai/v1/responses"
  api_key <- Sys.getenv("OPENAI_API_KEY")

  payload <- list(
    model = "gpt-5.2-pro",
    input = list(
      list(
        role = "user",
        content = list(
          list(
            type = "input_text",
            text = "这张图片里有什么？"
          ),
          list(
            type = "input_image",
            image_url = "https://openai-documentation.vercel.app/images/cat_and_otter.png"
          )
        )
      )
    )
  )

  response <- POST(
    url,
    add_headers(
      Authorization = paste("Bearer", api_key),
      `Content-Type` = "application/json"
    ),
    body = toJSON(payload, auto_unbox = TRUE),
    encode = "raw"
  )

  cat(content(response, "text"))
  ```
</RequestExample>

<ResponseExample>
  ```json 200 theme={null}
  {
    "code": 200,
    "data": {
      "id": "resp-9876543210",
      "object": "response",
      "created": 1677652288,
      "model": "gpt-5.2-pro",
      "choices": [
        {
          "index": 0,
          "message": {
            "role": "assistant",
            "content": "这张图片中有一只猫和一只水獭。它们看起来正在互动，场景非常可爱和温馨。猫咪和水獭似乎相处得很融洽。"
          },
          "finish_reason": "stop"
        }
      ],
      "usage": {
        "prompt_tokens": 156,
        "completion_tokens": 45,
        "total_tokens": 201
      }
    }
  }
  ```

  ```json 400 theme={null}
  {
    "error": {
      "code": 400,
      "message": "请求参数无效",
      "type": "invalid_request_error"
    }
  }
  ```

  ```json 401 theme={null}
  {
    "error": {
      "code": 401,
      "message": "身份验证失败，请检查您的API密钥",
      "type": "authentication_error"
    }
  }
  ```

  ```json 402 theme={null}
  {
    "error": {
      "code": 402,
      "message": "账户余额不足，请充值后再试",
      "type": "payment_required"
    }
  }
  ```

  ```json 403 theme={null}
  {
    "error": {
      "code": 403,
      "message": "访问被禁止，您没有权限访问此资源",
      "type": "permission_error"
    }
  }
  ```

  ```json 429 theme={null}
  {
    "error": {
      "code": 429,
      "message": "请求过于频繁，请稍后再试",
      "type": "rate_limit_error"
    }
  }
  ```

  ```json 500 theme={null}
  {
    "error": {
      "code": 500,
      "message": "服务器内部错误，请稍后重试",
      "type": "server_error"
    }
  }
  ```

  ```json 502 theme={null}
  {
    "error": {
      "code": 502,
      "message": "网关错误，服务器暂时不可用",
      "type": "bad_gateway"
    }
  }
  ```
</ResponseExample>

## Authorizations

<ParamField header="Authorization" type="string" required>
  所有接口均需要使用Bearer Token进行认证

  获取 API Key：

  访问 [API Key 管理页面](https://apimart.ai/keys) 获取您的 API Key

  使用时在请求头中添加：

  ```
  Authorization: Bearer YOUR_API_KEY
  ```
</ParamField>

## Body

<ParamField body="model" type="string" required default="gpt-5.2-pro">
  模型名称

  支持的模型包括：

  * `gpt-5.2-pro`
  * `gpt-5.2-codex`
  * 更多模型持续更新中...
</ParamField>

<ParamField body="input" type="array" required>
  输入内容列表

  输入数组，每个输入项包含 `role` 和 `content` 两个字段。

  **💡 快速填写（Try it 区域）：**

  1. 点击 "+ Add an item" 添加一个输入项
  2. `role` 输入：`user`（用户消息）、`assistant`（AI回复）或 `system`（系统提示词）
  3. `content` 添加内容块（可包含文本和图像）

  <Expandable title="详细字段说明">
    <ParamField body="role" type="string" required default="user">
      角色类型

      可选值：`user`（用户消息）、`assistant`（AI回复，用于多轮对话）、`system`（系统提示词，设置AI行为）
    </ParamField>

    <ParamField body="content" type="array" required>
      内容数组

      支持多种类型的内容块，可以包含文本和图像。

      <Expandable title="内容块类型">
        <ParamField body="type" type="string" required>
          内容类型

          可选值：

          * `input_text`: 文本输入
          * `input_image`: 图像输入
        </ParamField>

        <ParamField body="text" type="string">
          文本内容

          当 `type` 为 `input_text` 时使用，填写文本内容
        </ParamField>

        <ParamField body="image_url" type="string">
          图像URL

          当 `type` 为 `input_image` 时使用

          支持两种格式：

          **1. 完整的图像URL地址**

          * 公开可访问的图像URL（http\:// 或 https\://）
          * 示例：`https://example.com/image.jpg`

          **2. Base64 编码格式**

          * **必须使用完整的 Data URI 格式**
          * 格式：`data:image/{格式};base64,{base64数据}`
          * 支持的图片格式：jpeg、png、gif、webp
          * 示例：`data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEAYABg...`
          * ⚠️ 注意：必须包含 `data:image/jpeg;base64,` 前缀部分
        </ParamField>
      </Expandable>
    </ParamField>
  </Expandable>
</ParamField>

<ParamField body="temperature" type="number">
  控制输出随机性，范围 0-2

  * 较低的值（如 0.2）使输出更确定
  * 较高的值（如 1.8）使输出更随机

  默认值：1.0
</ParamField>

<ParamField body="max_tokens" type="integer">
  生成的最大token数量

  不同模型有不同的最大值限制，请参考具体模型文档
</ParamField>

<ParamField body="stream" type="boolean">
  是否使用流式输出

  * `true`: 流式返回（SSE格式）
  * `false`: 一次性返回完整响应

  默认值：false
</ParamField>

<ParamField body="top_p" type="number">
  核采样参数，范围 0-1

  控制生成文本的多样性，建议与 temperature 二选一使用

  默认值：1.0
</ParamField>

<ParamField body="tools" type="array">
  工具列表，用于扩展模型能力

  支持的工具类型：

  * **网络搜索** (`web_search`): 实时搜索互联网信息
  * **文件搜索** (`file_search`): 搜索已上传的文件内容
  * **函数调用** (`function`): 调用自定义函数
  * **远程MCP** (`remote_mcp`): 连接远程模型上下文协议服务

  示例：`[{"type": "web_search"}]`
</ParamField>

## Response

<ResponseField name="id" type="string">
  响应的唯一标识符
</ResponseField>

<ResponseField name="object" type="string">
  对象类型，固定为 `response`
</ResponseField>

<ResponseField name="created" type="integer">
  创建时间戳
</ResponseField>

<ResponseField name="model" type="string">
  实际使用的模型名称
</ResponseField>

<ResponseField name="choices" type="array">
  生成的回复列表

  <Expandable title="属性">
    <ResponseField name="index" type="integer">
      选项索引
    </ResponseField>

    <ResponseField name="message" type="object">
      消息内容

      <Expandable title="属性">
        <ResponseField name="role" type="string">
          角色类型（assistant）
        </ResponseField>

        <ResponseField name="content" type="string">
          生成的文本内容
        </ResponseField>
      </Expandable>
    </ResponseField>

    <ResponseField name="finish_reason" type="string">
      结束原因

      可能的值：

      * `stop` - 自然结束
      * `length` - 达到最大长度
      * `content_filter` - 内容过滤
    </ResponseField>
  </Expandable>
</ResponseField>

<ResponseField name="usage" type="object">
  token使用统计

  <Expandable title="属性">
    <ResponseField name="prompt_tokens" type="integer">
      输入内容的token数
    </ResponseField>

    <ResponseField name="completion_tokens" type="integer">
      生成内容的token数
    </ResponseField>

    <ResponseField name="total_tokens" type="integer">
      总token数
    </ResponseField>
  </Expandable>
</ResponseField>

## 使用示例

### 纯文本输入

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "input": [
    {
      "role": "user",
      "content": [
        {
          "type": "input_text",
          "text": "你好，介绍一下人工智能"
        }
      ]
    }
  ]
}
```

### 使用网络搜索工具

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "tools": [{"type": "web_search"}],
  "input": "今天有什么正面的新闻？"
}
```

```bash cURL示例 theme={null}
curl "https://api.apimart.ai/v1/responses" \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer <token>" \
    -d '{
        "model": "gpt-5.2-pro",
        "tools": [{"type": "web_search"}],
        "input": "今天有什么正面的新闻？"
    }'
```

### 图像理解

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "input": [
    {
      "role": "user",
      "content": [
        {
          "type": "input_text",
          "text": "描述这张图片"
        },
        {
          "type": "input_image",
          "image_url": "https://example.com/image.jpg"
        }
      ]
    }
  ]
}
```

### 多图像分析

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "input": [
    {
      "role": "user",
      "content": [
        {
          "type": "input_text",
          "text": "比较这两张图片的异同"
        },
        {
          "type": "input_image",
          "image_url": "https://example.com/image1.jpg"
        },
        {
          "type": "input_image",
          "image_url": "https://example.com/image2.jpg"
        }
      ]
    }
  ]
}
```

### Base64编码图像

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "input": [
    {
      "role": "user",
      "content": [
        {
          "type": "input_text",
          "text": "分析这张图片"
        },
        {
          "type": "input_image",
          "image_url": "data:image/jpeg;base64,/9j/4AAQSkZJRg..."
        }
      ]
    }
  ]
}
```

### 使用文件搜索工具

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "tools": [{"type": "file_search"}],
  "input": "根据已上传的文档，总结公司的季度业绩"
}
```

### 使用函数调用

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "获取指定城市的天气信息",
        "parameters": {
          "type": "object",
          "properties": {
            "city": {
              "type": "string",
              "description": "城市名称，例如：北京"
            },
            "unit": {
              "type": "string",
              "enum": ["celsius", "fahrenheit"],
              "description": "温度单位"
            }
          },
          "required": ["city"]
        }
      }
    }
  ],
  "input": "北京今天天气怎么样？"
}
```

### 使用远程MCP

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "tools": [
    {
      "type": "remote_mcp",
      "remote_mcp": {
        "url": "https://mcp.example.com/api",
        "auth_token": "your_mcp_token"
      }
    }
  ],
  "input": "查询数据库中的用户信息"
}
```

### 组合使用多个工具

```json theme={null}
{
  "model": "gpt-5.2-pro",
  "tools": [
    {"type": "web_search"},
    {"type": "file_search"},
    {
      "type": "function",
      "function": {
        "name": "calculate",
        "description": "执行数学计算",
        "parameters": {
          "type": "object",
          "properties": {
            "expression": {
              "type": "string",
              "description": "数学表达式"
            }
          },
          "required": ["expression"]
        }
      }
    }
  ],
  "input": "搜索最新的比特币价格，并计算100个比特币的总价值"
}
```

## 内容类型说明

### input\_text

文本输入类型

**属性：**

* `type`: 固定为 `"input_text"`
* `text`: 文本内容（字符串）

### input\_image

图像输入类型

**属性：**

* `type`: 固定为 `"input_image"`
* `image_url`: 图像URL或Base64编码的数据URI

**支持两种格式：**

1. **完整的图像URL地址**
   * 公开可访问的图像URL（http\:// 或 https\://）
   * 示例：`https://example.com/image.jpg`

2. **Base64 编码格式**
   * **必须使用完整的 Data URI 格式**
   * 格式：`data:image/{格式};base64,{base64数据}`
   * 示例：`data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEAYABg...`
   * ⚠️ 注意：必须包含 `data:image/jpeg;base64,` 前缀部分（其中 `jpeg` 可以替换为 `png`、`gif`、`webp` 等）

**支持的图像格式：**

* JPEG
* PNG
* GIF
* WebP

**图像大小限制：**

* 最大文件大小：20MB
* 推荐分辨率：不超过2048x2048像素

## 工具使用详解

### 网络搜索 (Web Search)

使用网络搜索工具可以让模型访问实时互联网信息。

**配置示例：**

```json theme={null}
{
  "tools": [{"type": "web_search"}]
}
```

**适用场景：**

* 查询最新新闻和时事
* 获取实时数据（股票、天气、汇率等）
* 搜索最新的技术文档和资料
* 验证事实信息

### 文件搜索 (File Search)

文件搜索工具允许模型在已上传的文档中搜索相关信息。

**配置示例：**

```json theme={null}
{
  "tools": [{"type": "file_search"}]
}
```

**适用场景：**

* 分析企业内部文档
* 搜索技术规范和手册
* 查询合同和法律文件
* 知识库问答系统

### 函数调用 (Function Calling)

定义自定义函数，让模型能够调用外部API或执行特定操作。

**完整配置示例：**

```json theme={null}
{
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_stock_price",
        "description": "获取股票的实时价格",
        "parameters": {
          "type": "object",
          "properties": {
            "symbol": {
              "type": "string",
              "description": "股票代码，例如：AAPL"
            },
            "currency": {
              "type": "string",
              "enum": ["USD", "CNY"],
              "description": "货币单位",
              "default": "USD"
            }
          },
          "required": ["symbol"]
        }
      }
    }
  ]
}
```

**参数说明：**

* `name`: 函数名称（必需）
* `description`: 函数功能描述（必需）
* `parameters`: 参数定义，使用JSON Schema格式
  * `type`: 参数类型
  * `properties`: 参数属性定义
  * `required`: 必需参数列表

**适用场景：**

* 调用第三方API
* 执行数据库查询
* 触发业务流程
* 与内部系统集成

### 远程MCP (Remote MCP)

连接到远程模型上下文协议（MCP）服务，扩展模型能力。

**配置示例：**

```json theme={null}
{
  "tools": [
    {
      "type": "remote_mcp",
      "remote_mcp": {
        "url": "https://your-mcp-server.com/api",
        "auth_token": "your_auth_token",
        "timeout": 30
      }
    }
  ]
}
```

**参数说明：**

* `url`: MCP服务器地址（必需）
* `auth_token`: 认证令牌（可选）
* `timeout`: 超时时间（秒），默认30秒

**适用场景：**

* 连接企业级AI服务
* 使用专业领域模型
* 访问受保护的数据源
* 分布式AI系统集成

## 工具响应格式

当模型使用工具时，响应格式会包含工具调用信息：

```json theme={null}
{
  "id": "resp-123456",
  "object": "response",
  "created": 1677652288,
  "model": "gpt-5.2-pro",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": null,
        "tool_calls": [
          {
            "id": "call_abc123",
            "type": "function",
            "function": {
              "name": "get_weather",
              "arguments": "{\"city\": \"北京\"}"
            }
          }
        ]
      },
      "finish_reason": "tool_calls"
    }
  ]
}
```

**工具调用流程：**

1. 模型接收用户输入
2. 分析是否需要使用工具
3. 如需要，返回工具调用请求
4. 客户端执行工具调用
5. 将工具结果返回给模型
6. 模型生成最终响应

## 注意事项

1. **图像URL要求**：
   * 必须是公开可访问的URL
   * 或使用Base64编码的Data URI格式

2. **Token计费**：
   * 图像会根据其分辨率消耗相应的tokens
   * 高分辨率图像会自动调整大小以优化成本
   * 工具调用也会消耗额外的tokens

3. **内容顺序**：
   * content数组中的元素顺序会影响模型理解
   * 建议先放置文本指令，再放置图像

4. **多模态组合**：
   * 可以在一个请求中混合多个文本和图像
   * 支持多轮对话，保持上下文连贯性

5. **工具使用限制**：
   * 同时使用多个工具时，模型会智能选择最合适的工具
   * 函数调用需要明确的函数定义和参数说明
   * 网络搜索结果可能受地域和时间限制

6. **API兼容性**：
   * 完全兼容OpenAI Responses API格式
   * 可无缝迁移现有OpenAI代码
   * 支持所有OpenAI工具扩展功能