跳到主要内容

发送第一个请求

只需将 base_url 替换为 Real200 的地址，即可使用与 OpenAI 完全相同的代码发送请求。

前置条件​

  • 已创建 API Key
  • 已安装 Python 3.8+ 或 Node.js 16+（或使用 cURL）

方式一：cURL（无需安装）​

curl https://real200.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $REAL200_API_KEY" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "你好，请介绍一下 Real200。"}]
  }'

方式二：Python（推荐）​

安装​

pip install openai

调用​

from openai import OpenAI

client = OpenAI(
    api_key="sk-real200-xxxxxxxxxxxxxxxx",  # 替换为您的 Real200 API Key
    base_url="https://real200.com/v1"       # 只需修改这一行
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "你好，请介绍一下 Real200。"}]
)

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

环境变量方式​

import os
from openai import OpenAI

# 设置环境变量（推荐）
os.environ["OPENAI_API_KEY"] = "sk-real200-xxxxxxxxxxxxxxxx"
os.environ["OPENAI_BASE_URL"] = "https://real200.com/v1"

# 自动读取环境变量
client = OpenAI()

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "你好！"}]
)

方式三：Node.js​

安装​

npm install openai

调用​

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-real200-xxxxxxxxxxxxxxxx", // 替换为您的 Real200 API Key
  baseURL: "https://real200.com/v1",     // 只需修改这一行
});

const response = await client.chat.completions.create({
  model: "gpt-4o",
  messages: [{ role: "user", content: "你好，请介绍一下 Real200。" }],
});

console.log(response.choices[0].message.content);

流式响应​

Real200 完全支持流式（streaming）响应，适合需要实时显示生成内容的场景。

from openai import OpenAI

client = OpenAI(
    api_key="sk-real200-xxxxxxxxxxxxxxxx",
    base_url="https://real200.com/v1"
)

stream = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "请写一首关于 AI 的诗。"}],
    stream=True  # 启用流式
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

响应示例​

成功响应（200 OK）​

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1716300000,
  "model": "gpt-4o",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Real200 是一个 AI Token 网关平台..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 12,
    "completion_tokens": 45,
    "total_tokens": 57
  }
}

流式响应片段​

{"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":"Real"},"finish_reason":null}]}
{"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":"200"},"finish_reason":null}]}
...
{"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

支持的模型​

您可以通过以下 API 查看 Real200 支持的模型列表：

curl https://real200.com/v1/models \
  -H "Authorization: Bearer $REAL200_API_KEY"

或在模型列表中查看。

下一步​

常见问题​

返回 401 Unauthorized 错误
  • 检查 API Key 是否正确（以 sk-real200- 开头）
  • 确认 Authorization: Bearer 格式正确
  • 确认 Key 未被禁用或删除
返回 429 Too Many Requests 错误
  • 您的请求频率超过了限流阈值
  • 查看限流策略了解具体限制
  • 等待片刻后重试，或联系客服提高配额
返回的模型不是我指定的

Real200 的智能路由可能将请求路由到等效的替代模型。如果您需要固定使用特定模型，请在渠道配置中关闭自动路由。