对话补全 Chat Completions

Chat Completions 是最核心的 API 端点，用于与大语言模型进行对话交互。

端点

POST https://api2everything.xyz/v1/chat/completions

请求时需在 Header 中携带 API Key 进行认证：

Authorization: Bearer sk-your-api-key
Content-Type: application/json

请求参数

Messages 格式

messages 是一个消息对象数组，每个对象包含 role 和 content 字段：

角色类型

content 字段

content 支持两种格式：

• 字符串 — 最常见的纯文本格式："content": "你好"
• 数组 — 用于多模态输入（如图片识别），数组中可包含文本和图片 URL：

{
  "role": "user",
  "content": [
    {"type": "text", "text": "这张图片里有什么？"},
    {"type": "image_url", "image_url": {"url": "https://example.com/photo.jpg"}}
  ]
}

多轮对话

要实现多轮对话，需要将之前的对话历史按顺序放入 messages 数组中。模型会根据完整的上下文生成回复：

{
  "model": "gpt-4o",
  "messages": [
    {"role": "system", "content": "你是一个有帮助的助手。"},
    {"role": "user", "content": "法国的首都是哪里？"},
    {"role": "assistant", "content": "法国的首都是巴黎。"},
    {"role": "user", "content": "那里有什么著名景点？"}
  ]
}

响应格式

成功调用后返回如下 JSON 结构：

{
  "id": "chatcmpl-abc123def456",
  "object": "chat.completion",
  "created": 1713345600,
  "model": "gpt-4o-2024-08-06",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "你好！有什么可以帮助你的吗？"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 15,
    "total_tokens": 27
  }
}

字段说明

示例

基本请求 — Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://api2everything.xyz/v1"
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "user", "content": "用一句话解释量子计算"}
    ],
    temperature=0.7,
    max_tokens=200
)

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

基本请求 — cURL

curl https://api2everything.xyz/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [
      {"role": "user", "content": "用一句话解释量子计算"}
    ],
    "temperature": 0.7,
    "max_tokens": 200
  }'

多轮对话 — Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://api2everything.xyz/v1"
)

# 维护对话历史
history = [
    {"role": "system", "content": "你是一个专业的烹饪助手。"}
]

def chat(user_message):
    history.append({"role": "user", "content": user_message})
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=history
    )
    reply = response.choices[0].message.content
    history.append({"role": "assistant", "content": reply})
    return reply

print(chat("我想做一道简单的意大利面"))
print(chat("家里没有培根，可以用什么替代？"))

使用 System Prompt — cURL

curl https://api2everything.xyz/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-20250514",
    "messages": [
      {"role": "system", "content": "你是一个资深的 Python 开发者，回答简洁且附带代码示例。"},
      {"role": "user", "content": "如何用 Python 读取 CSV 文件？"}
    ],
    "temperature": 0.3
  }'

错误处理

当请求失败时，API 返回包含 error 对象的 JSON 响应：

{
  "error": {
    "message": "Invalid API key provided.",
    "type": "authentication_error",
    "code": "invalid_api_key"
  }
}

常见错误码

重试建议

import time
from openai import OpenAI, APIError, RateLimitError

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://api2everything.xyz/v1"
)

def chat_with_retry(messages, max_retries=3):
    for attempt in range(max_retries):
        try:
            return client.chat.completions.create(
                model="gpt-4o",
                messages=messages
            )
        except RateLimitError:
            wait = 2 ** attempt  # 1s, 2s, 4s
            print(f"频率限制，{wait}秒后重试...")
            time.sleep(wait)
        except APIError as e:
            if e.status_code == 500 and attempt < max_retries - 1:
                time.sleep(2 ** attempt)
                continue
            raise
    raise Exception("重试次数已用尽")