跳到主要内容

cURL 调用

通过 HTTP API 直接调用 CloudBase AI 大模型，适用于后端服务、脚本、任意编程语言。

准备工作​

  1. 已创建云开发环境（旧套餐可升级），并已启用模型开关（见 接入大模型）
  2. 已创建 API Key（获取地址）

API 端点​

https://<ENV_ID>.api.tcloudbasegateway.com/v1/ai/<PROVIDER>/chat/completions
参数说明示例
ENV_ID环境 IDyour-env-id
PROVIDER模型提供商cloudbase

认证​

在请求头中添加 API Key：

Authorization: Bearer <YOUR_API_KEY>

文本生成​

非流式调用​

curl -X POST 'https://<ENV_ID>.api.tcloudbasegateway.com/v1/ai/cloudbase/chat/completions' \
  -H 'Authorization: Bearer <YOUR_API_KEY>' \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "hy3-preview",
    "messages": [
      {"role": "user", "content": "介绍一下李白"}
    ],
    "stream": false
  }'

响应示例：

{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "created": 1234567890,
  "model": "hy3-preview",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "李白（701年—762年），字太白，号青莲居士..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 150,
    "total_tokens": 160
  }
}

流式调用​

curl -X POST 'https://<ENV_ID>.api.tcloudbasegateway.com/v1/ai/cloudbase/chat/completions' \
  -H 'Authorization: Bearer <YOUR_API_KEY>' \
  -H 'Content-Type: application/json' \
  -H 'Accept: text/event-stream' \
  -d '{
    "model": "hy3-preview",
    "messages": [
      {"role": "user", "content": "介绍一下你自己"}
    ],
    "stream": true
  }'

响应格式（SSE）：

data: {"id":"chatcmpl-xxx","choices":[{"delta":{"content":"我"},"index":0}]}

data: {"id":"chatcmpl-xxx","choices":[{"delta":{"content":"是"},"index":0}]}

data: {"id":"chatcmpl-xxx","choices":[{"delta":{"content":"DeepSeek"},"index":0}]}

data: [DONE]

请求参数​

参数类型必填说明
modelstring是模型名称
messagesarray是消息列表
streamboolean否是否流式返回，默认 false
temperaturenumber否采样温度，范围 0-2
top_pnumber否核采样，范围 0-1
max_tokensnumber否最大生成 token 数

messages 格式​

[
  {"role": "system", "content": "你是一个有帮助的助手"},
  {"role": "user", "content": "用户问题"},
  {"role": "assistant", "content": "助手回复"},
  {"role": "user", "content": "后续问题"}
]
role说明
system系统提示，设定助手行为
user用户消息
assistant助手回复

多轮对话​

多轮对话需要在每次请求中携带完整的历史消息，详见多轮对话文档。

错误处理​

调用过程中遇到错误？请参考 常见错误码。

编程语言示例​

Python​

import requests

url = "https://<ENV_ID>.api.tcloudbasegateway.com/v1/ai/cloudbase/chat/completions"
headers = {
    "Authorization": "Bearer <YOUR_API_KEY>",
    "Content-Type": "application/json"
}
data = {
    "model": "hy3-preview",
    "messages": [{"role": "user", "content": "你好"}],
    "stream": False
}

response = requests.post(url, headers=headers, json=data)
result = response.json()
print(result["choices"][0]["message"]["content"])

Node.js​

const response = await fetch(
  "https://<ENV_ID>.api.tcloudbasegateway.com/v1/ai/cloudbase/chat/completions",
  {
    method: "POST",
    headers: {
      "Authorization": "Bearer <YOUR_API_KEY>",
      "Content-Type": "application/json"
    },
    body: JSON.stringify({
      model: "hy3-preview",
      messages: [{ role: "user", content: "你好" }]
    })
  }
);

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