ChatCompletions格式

ChatCompletions格式

POST /v1/chat/completions

根据对话历史创建模型响应。支持流式和非流式响应。

兼容 OpenAI Chat Completions API。

请求头 (Headers)

参数名	类型	必选	描述	示例
Authorization	string	否	鉴权令牌	Bearer ********************

请求体 (Request Body)

Content-Type: application/json

字段	类型	必选	描述
model	string	是	模型 ID
messages	array of any	是	对话消息列表
temperature	number	否	采样温度
top_p	number	否	核采样参数
n	integer	否	生成数量
stream	boolean	否	是否流式响应
stream_options	object	否
stream_options.include_usage	boolean	否
stop	any	否	停止序列
max_tokens	integer	否	最大生成 Token 数
max_completion_tokens	integer	否	最大补全 Token 数
presence_penalty	number	否
frequency_penalty	number	否
logit_bias	object	否
user	string	否
tools	array of any	否
tool_choice	any	否
response_format	any	否
seed	integer	否
reasoning_effort	string	否	推理强度 (用于支持推理的模型)
modalities	array of string	否
audio	object	否
audio.voice	string	否
audio.format	string	否

请求体示例

{
  "model": "gpt-4",
  "messages": [
    {
      "role": "string",
      "content": null,
      "name": "string",
      "tool_calls": [
        {
          "id": "string",
          "type": "function",
          "function": {
            "name": "string",
            "arguments": "string"
          }
        }
      ],
      "tool_call_id": "string",
      "reasoning_content": "string"
    }
  ],
  "temperature": 0,
  "top_p": 0,
  "n": 0,
  "stream": true,
  "stream_options": {
    "include_usage": true
  },
  "stop": null,
  "max_tokens": 0,
  "max_completion_tokens": 0,
  "presence_penalty": 0,
  "frequency_penalty": 0,
  "logit_bias": {},
  "user": "string",
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "string",
        "description": "string",
        "parameters": {}
      }
    }
  ],
  "tool_choice": null,
  "response_format": {
    "type": "string",
    "json_schema": {}
  },
  "seed": 0,
  "reasoning_effort": "string",
  "modalities": [
    "string"
  ],
  "audio": {
    "voice": "string",
    "format": "string"
  }
}

返回响应 (Responses)

200 - 成功创建响应

字段	类型	描述
id	string
object	string
created	integer
model	string
choices	array of object
choices[].index	integer
choices[].message	any
choices[].finish_reason	string
usage	any
system_fingerprint	string

响应示例

{
  "id": "string",
  "object": "chat.completion",
  "created": 0,
  "model": "string",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "string",
        "content": null,
        "name": "string",
        "tool_calls": [
          {
            "id": "string",
            "type": "function",
            "function": {
              "name": "string",
              "arguments": "string"
            }
          }
        ],
        "tool_call_id": "string",
        "reasoning_content": "string"
      },
      "finish_reason": "string"
    }
  ],
  "usage": {
    "prompt_tokens": 0,
    "completion_tokens": 0,
    "total_tokens": 0,
    "prompt_tokens_details": {
      "cached_tokens": 0,
      "text_tokens": 0,
      "audio_tokens": 0,
      "image_tokens": 0
    },
    "completion_tokens_details": {
      "text_tokens": 0,
      "audio_tokens": 0,
      "reasoning_tokens": 0
    }
  },
  "system_fingerprint": "string"
}

400 - 请求参数错误

字段	类型	描述
error	object
error.message	string	错误信息
error.type	string	错误类型
error.param	string (nullable)	相关参数
error.code	string (nullable)	错误代码

响应示例

{
  "error": {
    "message": "string",
    "type": "string",
    "param": "string",
    "code": "string"
  }
}

429 - 请求频率限制

字段	类型	描述
error	object
error.message	string	错误信息
error.type	string	错误类型
error.param	string (nullable)	相关参数
error.code	string (nullable)	错误代码

响应示例

{
  "error": {
    "message": "string",
    "type": "string",
    "param": "string",
    "code": "string"
  }
}