Chat 对话补全

OpenAI 兼容的对话补全接口，根据对话历史生成模型回复。支持流式和非流式响应。

参考文档: https://platform.openai.com/docs/api-reference/chat/create

请求

POST /v1/chat/completions

认证

使用 Bearer Token 认证，格式: Authorization: Bearer sk-xxxxxx

请求体 (application/json)

参数	类型	必填	说明
model	string	是	模型 ID（如 gpt-4）
messages	array[Message]	是	对话消息列表
temperature	number	否	采样温度，范围 0-2，默认 1
top_p	number	否	核采样参数，范围 0-1，默认 1
n	integer	否	生成补全数量，最小 1，默认 1
stream	boolean	否	是否流式响应，默认 false
stream_options	object	否	流式选项，包含 include_usage (boolean)
stop	string/array	否	停止序列
max_tokens	integer	否	最大生成 token 数
max_completion_tokens	integer	否	最大总 token 数
presence_penalty	number	否	存在惩罚，范围 -2 到 2，默认 0
frequency_penalty	number	否	频率惩罚，范围 -2 到 2，默认 0
logit_bias	object	否	token ID 到偏置值的映射
user	string	否	用户标识
tools	array[Tool]	否	模型可调用的工具列表
tool_choice	string/object	否	none、auto、required 或具体工具对象
response_format	object	否	格式规范（text、json_object、json_schema）
seed	integer	否	确定性采样种子
reasoning_effort	string	否	推理努力程度：low、medium、high
modalities	array[string]	否	模态：text、audio
audio	object	否	音频配置，含 voice 和 format

Message 对象

参数	类型	必填	说明
role	string	是	system、user、assistant、tool、developer
content	string/array	是	消息内容
name	string	否	参与者名称
tool_calls	array	否	工具调用（assistant 消息）
tool_call_id	string	否	工具调用 ID（tool 角色消息）
reasoning_content	string	否	推理内容

多模态消息内容 (MessageContent)

参数	类型	说明
type	string	text、image_url、input_audio、file、video_url
text	string	文本内容（type=text 时）
image_url	object	`{url: string, detail: "low"
input_audio	object	`{data: string(base64), format: "wav"
file	object	{filename, file_data, file_id}
video_url	object	{url: string}

请求示例

cURL

curl -X POST "https://your-api-base/v1/chat/completions" \
  -H "Authorization: Bearer sk-xxxxxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ],
    "temperature": 0.7,
    "stream": false
  }'

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-xxxxxx",
    base_url="https://your-api-base/v1"
)

response = client.chat.completions.create(
    model="gpt-4",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ],
    temperature=0.7
)

print(response.choices[0].message.content)

JavaScript

const response = await fetch("https://your-api-base/v1/chat/completions", {
  method: "POST",
  headers: {
    "Authorization": "Bearer sk-xxxxxx",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "gpt-4",
    messages: [
      { role: "system", content: "You are a helpful assistant." },
      { role: "user", content: "Hello!" }
    ],
    temperature: 0.7
  })
});

const data = await response.json();
console.log(data.choices[0].message.content);

响应

200 成功响应

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1677858242,
  "model": "gpt-4",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you today?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 20,
    "completion_tokens": 8,
    "total_tokens": 28,
    "prompt_tokens_details": {
      "cached_tokens": 0,
      "text_tokens": 20,
      "audio_tokens": 0,
      "image_tokens": 0
    },
    "completion_tokens_details": {
      "text_tokens": 8,
      "audio_tokens": 0,
      "reasoning_tokens": 0
    }
  },
  "system_fingerprint": "fp_44709d6fcb"
}

流式响应 (chunk)

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion.chunk",
  "created": 1677858242,
  "model": "gpt-4",
  "choices": [
    {
      "index": 0,
      "delta": {
        "role": "assistant",
        "content": "Hello"
      },
      "finish_reason": null
    }
  ]
}

错误响应

{
  "error": {
    "message": "Invalid API key",
    "type": "invalid_request_error",
    "param": null,
    "code": "invalid_api_key"
  }
}