### API Response Structure

Source: https://dots.ai/platform/docs

Example of a successful chat completion response object returned by the API.

```json
  "model": "dots3-note-prev",
  "object": "chat.completion",
  "system_fingerprint": "",
  "usage": {
    "prompt_tokens": 1291,
    "completion_tokens": 1746,
    "total_tokens": 3037
  }
}
```

--------------------------------

### Audio Transcription API Response

Source: https://dots.ai/platform/docs

Example response body containing the transcribed text and reasoning process from the model.

```json
{
  "id": "3ea79cea726648b9bb37ed72338e3965",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "How old is the Brooklyn Bridge?",
        "reasoning_content": "Thinking Process:\n\n1.  **Analyze the Request:** The user wants a transcription of the audio file. The output must be *only* the transcription text, with no extra analysis, explanation, prefix, or suffix.\n2.  **Listen to the Audio:** The audio says \"How old is the Brooklyn Bridge?\"\n3.  **Transcribe the Audio:** \"How old is the Brooklyn Bridge?\"\n4.  **Format the Output:** Just the transcribed text. \"How old is the Brooklyn Bridge?\"\n"
      }
    }
  ],
  "created": 1786627279,
  "model": "dots3-note-prev",
  "object": "chat.completion",
  "system_fingerprint": "",
  "usage": {
    "prompt_tokens": 55,
    "completion_tokens": 116,
    "total_tokens": 171
  }
}
```

--------------------------------

### 基础文本对话调用

Source: https://dots.ai/platform/docs

通过 cURL 发送非流式文本请求，系统指令通过顶层 system 参数传入。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/messages \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "dots3-note-prev",
    "system": "你是一个专业、可靠的 AI 助手。",
    "messages": [
      {
        "role": "user",
        "content": "你好，请介绍一下自己"
      }
    ],
    "max_tokens": 1024,
    "stream": false,
    "thinking": {
      "type": "disabled"
    }
  }'
```

--------------------------------

### 调用 OpenAI 聊天补全接口

Source: https://dots.ai/platform/docs

使用 curl 发起基础文本对话请求，需在请求头中携带有效的 api-key。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "system",
        "content": "你是一个AI助手"
      },
      {
        "role": "user",
        "content": "你好，请介绍一下自己"
      }
    ],
    "stream": false,
    "max_tokens": 1024,
    "temperature": 0.9
  }'
```

--------------------------------

### 发送工具调用请求

Source: https://dots.ai/platform/docs

使用 curl 发送包含 tools 定义的 POST 请求，模型将根据描述判断是否调用函数。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "type": "text",
            "text": "波士顿今天天气怎么样？请使用 get_current_weather 工具查询。"
          }
        ]
      }
    ],
    "tools": [
      {
        "type": "function",
        "function": {
          "name": "get_current_weather",
          "description": "查询指定地点的当前天气",
          "parameters": {
            "type": "object",
            "properties": {
              "city": {
                "type": "string",
                "description": "城市名称，例如 Boston"
              },
              "state": {
                "type": "string",
                "description": "州名的两个字母缩写，例如 MA"
              },
              "unit": {
                "type": "string",
                "description": "温度单位",
                "enum": ["celsius"]
              }
            },
            "required": ["city", "state", "unit"]
          }
        }
      }
    ],
    "stream": false,
    "max_tokens": 1024,
    "temperature": 0.9
  }'
```

--------------------------------

### 启用深度思考的 API 请求

Source: https://dots.ai/platform/docs

通过设置 chat_template_kwargs.enable_thinking 为 true 显式开启深度思考功能。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "system",
        "content": "你是一个AI助手"
      },
      {
        "role": "user",
        "content": "你好，请介绍一下自己"
      }
    ],
    "stream": false,
    "max_tokens": 1024,
    "temperature": 0.9,
    "chat_template_kwargs": {
      "enable_thinking": true
    }
  }'
```

--------------------------------

### 处理工具调用响应

Source: https://dots.ai/platform/docs

当 finish_reason 为 tool_calls 时，从 message.tool_calls 中获取函数名称及参数进行执行。

```json
{
  "id": "1f79f7638fa145de844a0851a892d01f",
  "choices": [
    {
      "index": 0,
      "finish_reason": "tool_calls",
      "message": {
        "role": "assistant",
        "reasoning_content": "用户想查询波士顿的天气。我们需要调用 get_current_weather 工具，传入 city=\"Boston\" 和 state=\"MA\"（马萨诸塞州的缩写）。单位使用摄氏度。\n",
        "tool_calls": [
          {
            "index": 0,
            "type": "function",
            "id": "call_9ef22900ae524ce4a94120a0",
            "function": {
              "name": "get_current_weather",
              "arguments": "{\"city\": \"Boston\", \"state\": \"MA\", \"unit\": \"celsius\"}"
            }
          }
        ]
      }
    }
  ],
  "created": 1786528610,
  "model": "dots3-note-prev",
  "object": "chat.completion",
  "system_fingerprint": "",
  "usage": {
    "prompt_tokens": 246,
    "completion_tokens": 99,
    "total_tokens": 345
  }
}
```

--------------------------------

### 提交工具执行结果

Source: https://dots.ai/platform/docs

在后续请求中包含完整的对话历史，并将 tool_result 放入消息数组中以获取最终回复。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/messages \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "user",
        "content": "波士顿今天天气怎么样？请使用 get_current_weather 工具查询。"
      },
      {
        "role": "assistant",
        "content": [
          {
            "type": "text",
            "text": "我需要使用 get_current_weather 工具来查询波士顿的天气，但该工具需要提供州（state）参数。波士顿位于马萨诸塞州（Massachusetts），因此我将使用该信息进行查询。"
          },
          {
            "type": "tool_use",
            "id": "call_8bf8d5e73316496b9d9181e0",
            "name": "get_current_weather",
            "input": {
              "city": "Boston",
              "state": "Massachusetts",
              "unit": "celsius"
            }
          }
        ]
      },
      {
        "role": "user",
        "content": [
          {
            "type": "tool_result",
            "tool_use_id": "call_8bf8d5e73316496b9d9181e0",
            "content": "当前温度 24°C，晴，湿度 42%"
          }
        ]
      }
    ],
    "tools": [
      {
        "name": "get_current_weather",
        "description": "查询指定地点的当前天气",
        "input_schema": {
          "type": "object",
          "properties": {
            "city": {
              "type": "string"
            },
            "state": {
              "type": "string"
            },
            "unit": {
              "type": "string",
              "enum": [
                "celsius"
              ]
            }
          },
          "required": [
            "city",
            "state",
            "unit"
          ]
        }
      }
    ],
    "thinking": {
      "type": "disabled"
    },
    "max_tokens": 1024,
    "stream": false
  }'
```

--------------------------------

### 发送工具定义请求

Source: https://dots.ai/platform/docs

通过 POST 请求发送包含工具定义的初始消息，模型将根据输入决定是否调用工具。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/messages \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "user",
        "content": "波士顿今天天气怎么样？请使用 get_current_weather 工具查询。"
      }
    ],
    "tools": [
      {
        "name": "get_current_weather",
        "description": "查询指定地点的当前天气",
        "input_schema": {
          "type": "object",
          "properties": {
            "city": {
              "type": "string"
            },
            "state": {
              "type": "string"
            },
            "unit": {
              "type": "string",
              "enum": [
                "celsius"
              ]
            }
          },
          "required": [
            "city",
            "state",
            "unit"
          ]
        }
      }
    ],
    "tool_choice": {
      "type": "auto"
    },
    "max_tokens": 1024,
    "stream": false,
    "thinking": {
      "type": "disabled"
    }
  }'
```

--------------------------------

### 视频输入 API 调用示例

Source: https://dots.ai/platform/docs

通过 API 传入视频 URL 进行分析。注意视频处理的 TTFT 可能较长，建议客户端设置合理的超时时间。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "system",
        "content": [
          {
            "type": "text",
            "text": "我是你的生活助手。"
          }
        ]
      },
      {
        "role": "user",
        "content": [
          {
            "type": "image_url",
            "image_url": {
              "url": "https://ci.xiaohongshu.com/1040g2sg31stpngqdlm6g5qbe3vodp7j07rja8mg?imageView2/2/w/1080/format/jpg"
            }
          },
          {
            "type": "video_url",
            "video_url": {
              "url": "https://avtshare01.rz.tu-ilmenau.de/avt-vqdb-uhd-1/test_1/segments/bigbuck_bunny_8bit_200kbps_360p_60.0fps_h264.mp4"
            }
          },
          {
            "type": "text",
            "text": "图片和视频有什么关联？"
          }
        ]
      }
    ],
    "stream": false,
    "max_tokens": 4096,
    "temperature": 0.9,
    "chat_template_kwargs": {
      "enable_thinking": true
    }
  }'
```

--------------------------------

### Configure Deep Thinking via API Request

Source: https://dots.ai/platform/docs

Use this curl command to send a request with adaptive thinking enabled. Ensure the api-key header is replaced with your actual key.

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/messages \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "user",
        "content": "请计算 27 × 43，并简要说明计算过程。"
      }
    ],
    "thinking": {
      "type": "adaptive"
    },
    "output_config": {
      "effort": "medium"
    },
    "max_tokens": 4096,
    "stream": false
  }'
```

--------------------------------

### 关闭深度思考的 API 请求

Source: https://dots.ai/platform/docs

通过设置 chat_template_kwargs.enable_thinking 为 false 关闭深度思考功能。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "system",
        "content": "你是一个AI助手"
      },
      {
        "role": "user",
        "content": "你好，请介绍一下自己"
      }
    ],
    "stream": false,
    "max_tokens": 1024,
    "temperature": 0.9,
    "chat_template_kwargs": {
      "enable_thinking": false
    }
  }'
```

--------------------------------

### 图片输入调用

Source: https://dots.ai/platform/docs

通过 cURL 发送包含图片 URL 的请求，图片与文本指令在同一内容数组中。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/messages \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "user",
        "content": [
          {
            "type": "image",
            "source": {
              "type": "url",
              "url": "https://picasso-static.xiaohongshu.com/fe-platform/e6d7a4f5dd20a3574caaa72b2b68217bc2e5e2e7.png"
            }
          },
          {
            "type": "text",
            "text": "请用一句话描述这张图片。"
          }
        ]
      }
    ],
    "max_tokens": 2048,
    "stream": false,
    "thinking": {
      "type": "disabled"
    }
  }'
```

--------------------------------

### 图片输入 API 调用示例

Source: https://dots.ai/platform/docs

通过 POST 请求发送包含文本和图片 URL 的多模态消息。请确保替换示例中的 API Key 为实际值。

```bash
curl -X POST https://note3-prev-api.askdiandian.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "system",
        "content": [
          {
            "type": "text",
            "text": "我是你的生活助手。"
          }
        ]
      },
      {
        "role": "user",
        "content": [
          {
            "type": "text",
            "text": "请描述这张图片。"
          },
          {
            "type": "image_url",
            "image_url": {
              "url": "https://picasso-static.xiaohongshu.com/fe-platform/e6d7a4f5dd20a3574caaa72b2b68217bc2e5e2e7.png",
              "detail": "medium"
            }
          }
        ]
      }
    ],
    "stream": false,
    "max_tokens": 1024,
    "temperature": 0.9,
    "chat_template_kwargs": {
      "enable_thinking": false
    }
  }'
```

--------------------------------

### 工具调用响应示例

Source: https://dots.ai/platform/docs

模型返回包含 tool_use 块的响应，指示需要执行的工具名称及参数。

```json
{
  "id": "msg_a61a2da6c80e438897932c37ea3f16a1",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "我需要使用 get_current_weather 工具来查询波士顿的天气，但该工具需要提供州（state）参数。波士顿位于马萨诸塞州（Massachusetts），因此我将使用该信息进行查询。"
    },
    {
      "type": "tool_use",
      "id": "call_8bf8d5e73316496b9d9181e0",
      "name": "get_current_weather",
      "input": {
        "city": "Boston",
        "state": "Massachusetts",
        "unit": "celsius"
      }
    }
  ],
  "model": "dots3-note-prev",
  "stop_reason": "tool_use",
  "usage": {
    "input_tokens": 219,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 100
  }
}
```

--------------------------------

### POST /v1/chat/completions

Source: https://dots.ai/platform/docs

根据输入的消息列表生成模型回复，支持文本、图片、视频、音频输入以及工具调用和深度思考功能。

```APIDOC
## POST /v1/chat/completions

### Description
根据输入的消息列表生成模型回复。当前 `dots3-note-prev` 的上下文窗口为 512K Token。接口支持文本、图片、视频和音频输入，以及工具调用、深度思考和流式响应。

### Method
POST

### Endpoint
https://note3-prev-api.askdiandian.com/v1/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - 要调用的模型名称，当前示例为 dots3-note-prev。
- **messages** (array) - Required - 按顺序排列的对话消息列表。
- **temperature** (number) - Optional - 控制输出随机性。
- **top_p** (number) - Optional - 控制核采样的候选 Token 范围。
- **stream** (boolean) - Optional - 是否以流式方式返回结果。
- **max_tokens** (integer) - Optional - 本次生成允许使用的最大输出 Token 数。
- **stop** (array) - Optional - 指定一个或多个停止序列。
- **tools** (array) - Optional - 当前请求可供模型选择的工具列表。
- **tool_choice** (string/object) - Optional - 控制模型是否以及如何选择工具。
- **chat_template_kwargs** (object) - Optional - 模型生成行为的可选配置对象。

### Request Example
{
  "model": "dots3-note-prev",
  "messages": [
    { "role": "system", "content": "你是一个AI助手" },
    { "role": "user", "content": "你好，请介绍一下自己" }
  ],
  "stream": false,
  "max_tokens": 1024,
  "temperature": 0.9
}

### Response
#### Success Response (200)
- **id** (string) - 响应唯一标识符
- **choices** (array) - 模型生成的回复列表
- **usage** (object) - Token 使用统计信息

#### Response Example
{
  "id": "bb04ce6f700f4b748c5819aa4a5eed8c",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "你好！我是 dots...",
        "reasoning_content": "..."
      }
    }
  ],
  "usage": {
    "prompt_tokens": 13,
    "completion_tokens": 155,
    "total_tokens": 168
  }
}
```

--------------------------------

### 流式响应请求示例

Source: https://dots.ai/platform/docs

使用 curl 发送带有 stream: true 参数的 POST 请求以获取流式响应。

```bash
curl -N -X POST https://note3-prev-api.askdiandian.com/v1/messages \
  -H "Content-Type: application/json" \
  -H "api-key: YOUR_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "dots3-note-prev",
    "messages": [
      {
        "role": "user",
        "content": "用一句话介绍 Dots。"
      }
    ],
    "max_tokens": 256,
    "stream": true,
    "thinking": {
      "type": "disabled"
    }
  }'
```

--------------------------------

### 解析聊天补全响应

Source: https://dots.ai/platform/docs

接口返回的完整 JSON 响应示例，包含模型回复内容及 Token 使用情况。

```json
{
  "id": "bb04ce6f700f4b748c5819aa4a5eed8c",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "你好！我是 dots，中文名叫点点，是小红书 Dots Studio 开发的 AI 大模型。在我们的设计理念里，“点”代表微小的个体。「点」也蕴含了 AI 串联每一个人、连接生活与知识的意味。我可以陪你聊天、帮你分析图片和文件，也能查资料、写代码。希望我能成为你身边自然、可靠又温暖的伙伴。",
        "reasoning_content": "用户问了一个简单的自我介绍问题。我需要简洁自然地介绍自己—— dots，可以 Mention 我是小红书 Dots Studio 开发的 AI 大模型，名字的含义，以及我存在的定位。不要堆砌背景，保持温暖友好。用户没有指定语言，我会用中文回答。\n"
      }
    }
  ],
  "created": 1786528606,
  "model": "dots3-note-prev",
  "object": "chat.completion",
  "system_fingerprint": "",
  "usage": {
    "prompt_tokens": 13,
    "completion_tokens": 155,
    "total_tokens": 168
  }
}
```

--------------------------------

### POST /v1/chat/completions

Source: https://dots.ai/platform/docs

Creates a chat completion response using the dots3-note-prev model, with support for enabling or disabling deep thinking.

```APIDOC
## POST /v1/chat/completions

### Description
Sends a message to the AI model to receive a completion. Supports deep thinking configuration via the chat_template_kwargs parameter.

### Method
POST

### Endpoint
https://note3-prev-api.askdiandian.com/v1/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model identifier, e.g., "dots3-note-prev"
- **messages** (array) - Required - List of message objects with role and content
- **stream** (boolean) - Optional - Whether to stream the response
- **max_tokens** (integer) - Optional - Maximum number of tokens to generate
- **temperature** (number) - Optional - Sampling temperature
- **chat_template_kwargs** (object) - Optional - Configuration object for model generation behavior
- **chat_template_kwargs.enable_thinking** (boolean) - Optional - Controls whether to enable deep thinking. Defaults to true.

### Request Example
{
  "model": "dots3-note-prev",
  "messages": [
    {"role": "system", "content": "你是一个AI助手"},
    {"role": "user", "content": "你好，请介绍一下自己"}
  ],
  "stream": false,
  "max_tokens": 1024,
  "temperature": 0.9,
  "chat_template_kwargs": {
    "enable_thinking": true
  }
}

### Response
#### Success Response (200)
- **id** (string) - Unique identifier for the completion
- **choices** (array) - List of completion choices
- **message** (object) - The assistant's message containing content and reasoning_content
- **usage** (object) - Token usage statistics

#### Response Example
{
  "id": "5656f21c7eca48cd86e752cf2e79b16b",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "你好！我是 dots...",
        "reasoning_content": "用户问的是自我介绍..."
      }
    }
  ],
  "model": "dots3-note-prev",
  "usage": {
    "prompt_tokens": 13,
    "completion_tokens": 118,
    "total_tokens": 131
  }
}
```

--------------------------------

### POST /v1/chat/completions

Source: https://dots.ai/platform/docs

Sends a chat request to the model with support for multimodal content such as audio or video URLs and text instructions.

```APIDOC
## POST /v1/chat/completions

### Description
Generates a chat completion response based on the provided messages, which can include text, audio, or video content.

### Method
POST

### Endpoint
https://note3-prev-api.askdiandian.com/v1/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model identifier, e.g., "dots3-note-prev".
- **messages** (array) - Required - An array of message objects containing role and content.
- **messages[].content** (array) - Required - Multimodal content array (text, audio_url, video_url).
- **stream** (boolean) - Optional - Whether to stream the response.
- **max_tokens** (integer) - Optional - Maximum number of tokens to generate.
- **temperature** (float) - Optional - Sampling temperature.
- **chat_template_kwargs** (object) - Optional - Additional configuration, such as `enable_thinking`.

### Request Example
{
  "model": "dots3-note-prev",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "audio_url",
          "audio_url": { "url": "https://example.com/audio.mp3" }
        },
        { "type": "text", "text": "Please transcribe this." }
      ]
    }
  ],
  "stream": false,
  "chat_template_kwargs": { "enable_thinking": true }
}

### Response
#### Success Response (200)
- **id** (string) - Unique identifier for the completion.
- **choices** (array) - List of completion choices.
- **usage** (object) - Token usage statistics.

#### Response Example
{
  "id": "3ea79cea726648b9bb37ed72338e3965",
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "How old is the Brooklyn Bridge?"
      }
    }
  ],
  "usage": {
    "prompt_tokens": 55,
    "completion_tokens": 116,
    "total_tokens": 171
  }
}
```

--------------------------------

### 基础文本对话响应

Source: https://dots.ai/platform/docs

上述基础文本请求返回的完整 JSON 响应示例。

```json
{
  "id": "msg_e6c1458beb264e458d078e0f29ea42dc",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "你好！我是 dots，由小红书 Dots Studio 开发，是一个 AI 大模型。很高兴认识你，有什么我可以帮你的吗？"
    }
  ],
  "model": "dots3-note-prev",
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 24,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 32
  }
}
```

--------------------------------

### POST /v1/chat/completions

Source: https://dots.ai/platform/docs

Sends a chat completion request with tool definitions to allow the model to perform function calling.

```APIDOC
## POST /v1/chat/completions

### Description
This endpoint allows the model to select and request the execution of defined tools (functions) based on user input.

### Method
POST

### Endpoint
https://note3-prev-api.askdiandian.com/v1/chat/completions

### Parameters
#### Request Body
- **model** (string) - Required - The model identifier.
- **messages** (array) - Required - The conversation history.
- **tools** (array) - Optional - A list of tools the model may call.
- **tool_choice** (string/object) - Optional - Controls tool calling behavior (none, auto, required, or specific function).
- **stream** (boolean) - Optional - Whether to stream the response.
- **max_tokens** (integer) - Optional - Maximum tokens to generate.
- **temperature** (number) - Optional - Sampling temperature.

### Request Example
{
  "model": "dots3-note-prev",
  "messages": [
    {
      "role": "user",
      "content": [{"type": "text", "text": "波士顿今天天气怎么样？请使用 get_current_weather 工具查询。"}]
    }
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_current_weather",
        "description": "查询指定地点的当前天气",
        "parameters": {
          "type": "object",
          "properties": {
            "city": {"type": "string", "description": "城市名称"},
            "state": {"type": "string", "description": "州名缩写"},
            "unit": {"type": "string", "description": "温度单位"}
          },
          "required": ["city", "state", "unit"]
        }
      }
    }
  ]
}

### Response
#### Success Response (200)
- **choices** (array) - The list of completion choices, containing tool_calls if triggered.
- **usage** (object) - Token usage statistics.

#### Response Example
{
  "choices": [
    {
      "finish_reason": "tool_calls",
      "message": {
        "tool_calls": [
          {
            "type": "function",
            "function": {
              "name": "get_current_weather",
              "arguments": "{\"city\": \"Boston\", \"state\": \"MA\", \"unit\": \"celsius\"}"
            }
          }
        ]
      }
    }
  ]
}
```

--------------------------------

### SSE 事件响应示例

Source: https://dots.ai/platform/docs

展示了流式接口返回的实际 SSE 事件序列，包括消息开始、内容块增量及结束事件。

```text
event: message_start
data: {"type":"message_start","message":{"id":"msg_d51dff47f30340d7b840525d036cba4c","type":"message","role":"assistant","content":[],"model":"dots3-note-prev","usage":{"input_tokens":22,"cache_creation_input_tokens":0,"cache_read_input_tokens":0,"cache_creation":{"ephemeral_5m_input_tokens":0,"ephemeral_1h_input_tokens":0},"output_tokens":0}}}

event: content_block_start
data: {"type":"content_block_start","index":0,"content_block":{"type":"text","text":""}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"D","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"ots","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":" 是由","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"小红书 D","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"ots Studio 开发","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"的 AI","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":" 大模型，","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"致力于提供","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"自然","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"、智能且贴近","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"生活的","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"对话与多","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"模态交互体验","stop_sequence":null}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"。","stop_sequence":null}}

event: content_block_stop
data: {"type":"content_block_stop","index":0}

event: message_delta
data: {"type":"message_delta","delta":{"stop_reason":"end_turn","stop_sequence":null},"usage":{"input_tokens":0,"cache_creation_input_tokens":0,"cache_read_input_tokens":0,"cache_creation":{"ephemeral_5m_input_tokens":0,"ephemeral_1h_input_tokens":0},"output_tokens":35}}

event: message_stop
data: {"type":"message_stop"}
```

--------------------------------

### 深度思考响应示例

Source: https://dots.ai/platform/docs

展示包含 reasoning_content 字段的 API 响应结构。

```json
{
  "id": "5656f21c7eca48cd86e752cf2e79b16b",
  "choices": [
    {
      "index": 0,
      "finish_reason": "stop",
      "message": {
        "role": "assistant",
        "content": "你好！我是 dots，中文名叫点点，是小红书 Dots Studio 开发的 AI 大模型。我可以帮你解答问题、梳理思路、分析图片或文件，也能陪你聊天、一起探讨生活里的各种小事。今天是想聊点什么呢？",
        "reasoning_content": "用户问的是自我介绍，需要简洁自然地说明我的身份、由谁开发、能做什么，同时保持友好和有温度的语气。不需要列一大串功能清单，重点突出核心能力和陪伴感。最后可以自然地反问一句，让对话活起来。\n"
      }
    }
  ],
  "created": 1786528611,
  "model": "dots3-note-prev",
  "object": "chat.completion",
  "system_fingerprint": "",
  "usage": {
    "prompt_tokens": 13,
    "completion_tokens": 118,
    "total_tokens": 131
  }
}
```

--------------------------------

### POST /v1/messages

Source: https://dots.ai/platform/docs

Sends a message to the model with tool definitions to initiate a tool call or provide tool results.

```APIDOC
## POST /v1/messages

### Description
Sends a request to the model. When tools are provided, the model may return a `tool_use` block. After executing the tool, the caller sends a `tool_result` back to the model to receive the final answer.

### Method
POST

### Endpoint
https://note3-prev-api.askdiandian.com/v1/messages

### Request Body
- **model** (string) - Required - The model identifier (e.g., "dots3-note-prev")
- **messages** (array) - Required - The conversation history including user and assistant roles
- **tools** (array) - Optional - Definitions of tools available to the model
- **tool_choice** (object) - Optional - Configuration for tool selection behavior
- **max_tokens** (integer) - Optional - Maximum tokens for the response
- **stream** (boolean) - Optional - Whether to stream the response

### Request Example
{
  "model": "dots3-note-prev",
  "messages": [
    {
      "role": "user",
      "content": "波士顿今天天气怎么样？请使用 get_current_weather 工具查询。"
    }
  ],
  "tools": [
    {
      "name": "get_current_weather",
      "description": "查询指定地点的当前天气",
      "input_schema": {
        "type": "object",
        "properties": {
          "city": { "type": "string" },
          "state": { "type": "string" },
          "unit": { "type": "string", "enum": ["celsius"] }
        },
        "required": ["city", "state", "unit"]
      }
    }
  ]
}

### Response
#### Success Response (200)
- **id** (string) - Unique message identifier
- **content** (array) - List of content blocks (text or tool_use)
- **stop_reason** (string) - Reason for stopping (e.g., "tool_use" or "end_turn")

#### Response Example
{
  "id": "msg_a61a2da6c80e438897932c37ea3f16a1",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "tool_use",
      "id": "call_8bf8d5e73316496b9d9181e0",
      "name": "get_current_weather",
      "input": { "city": "Boston", "state": "Massachusetts", "unit": "celsius" }
    }
  ],
  "stop_reason": "tool_use"
}
```

--------------------------------

### POST /v1/messages

Source: https://dots.ai/platform/docs

Sends a message to the model with optional tool definitions and deep thinking configuration.

```APIDOC
## POST /v1/messages

### Description
Sends a message to the model. Supports tool definitions for function calling and deep thinking configurations for reasoning tasks.

### Method
POST

### Endpoint
https://note3-prev-api.askdiandian.com/v1/messages

### Request Body
- **model** (string) - Required - The model identifier (e.g., "dots3-note-prev").
- **messages** (array) - Required - List of message objects.
- **thinking** (object) - Optional - Configuration for deep thinking (e.g., {"type": "adaptive"}).
- **output_config** (object) - Optional - Configuration for output effort (e.g., {"effort": "medium"}).
- **max_tokens** (integer) - Optional - Maximum number of tokens to generate.
- **stream** (boolean) - Optional - Whether to stream the response.
- **tools** (array) - Optional - List of tool definitions.
  - **name** (string) - Required - Tool name.
  - **description** (string) - Optional - Tool purpose.
  - **input_schema** (object) - Required - JSON Schema for tool input.

### Request Example
{
  "model": "dots3-note-prev",
  "messages": [
    {
      "role": "user",
      "content": "请计算 27 × 43，并简要说明计算过程。"
    }
  ],
  "thinking": {
    "type": "adaptive"
  },
  "output_config": {
    "effort": "medium"
  },
  "max_tokens": 4096,
  "stream": false
}

### Response
#### Success Response (200)
- **id** (string) - Unique message identifier.
- **type** (string) - Response type.
- **role** (string) - Role of the sender.
- **content** (array) - List of content blocks (thinking or text).
- **model** (string) - Model used.
- **usage** (object) - Token usage statistics.

#### Response Example
{
  "id": "msg_4352cea7df6a42859f37129b8b2fcafd",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "thinking",
      "thinking": "计算 27 × 43 可以拆成 (20 + 7) × (40 + 3)..."
    },
    {
      "type": "text",
      "text": "27 × 43 = 1161。"
    }
  ],
  "model": "dots3-note-prev",
  "usage": {
    "input_tokens": 27,
    "output_tokens": 228
  }
}
```

--------------------------------

### 获取基础 URL

Source: https://dots.ai/platform/docs

Dots API 的基础访问地址。

```text
https://note3-prev-api.askdiandian.com
```