首页 Home 产品 Product 文档 Docs 定价 Pricing 关于 About
登录 / 注册 Login / Sign Up
中文 English

开发文档Documentation

快速接入,立即开始使用大模型服务

Quick Integration, Start Using LLM Services Now

快速开始

Quick Start

1

注册账号

Register Account

Register Account

前往控制台注册账号,获取专属 API Key

Register at the console, get your API Key

2

选择模型

Choose Model

Choose Model

从已上线的模型中选择,发起 API 请求

Choose from available models, make API requests

3

按需充值

Recharge

Recharge

查看用量统计,余额不足时及时充值

Check usage stats, recharge when needed

API 基础信息

API Base Info

Base URL: http://47.99.105.150/v1

Base URL: http://47.99.105.150/v1

兼容性: 兼容 OpenAI API 格式,支持 OpenAI SDK 直接调用

Compatibility: OpenAI API compatible, use OpenAI SDK directly

聊天补全接口

Chat Completions API

POST /v1/chat/completions

POST /v1/chat/completions

请求参数

Request Parameters

参数名Param 类型Type 必填Required 说明Description
model string Yes 模型标识Model ID
messages array Yes 消息数组Message array
stream boolean No 流式返回Stream response
temperature float No 采样温度Sampling temperature
max_tokens integer No 最大Token数Max tokens

请求示例

Request Examples

# cURL
curl http://47.99.105.150/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "Doubao-pro-128k",
    "messages": [
      {"role": "user", "content": "你好Hello"}
    ]
  }'
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="http://47.99.105.150/v1"
)

response = client.chat.completions.create(
    model="Doubao-pro-128k",
    messages=[{"role": "user", "content": "你好Hello"}]
)
print(response.choices[0].message.content)

响应示例

Response Example

{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "created": 1700000000,
  "model": "Doubao-pro-128k",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "你好!有什么可以帮助你的吗?Hello! How can I help you?"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 20,
    "total_tokens": 30
  }
}

模型列表接口

List Models API

GET /v1/models

GET /v1/models

# cURL
curl http://47.99.105.150/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"

错误码说明

Error Codes

401 Unauthorized
认证失败,请检查 API Key 是否正确
Authentication failed, check your API Key
403 Forbidden
无权限访问,账户余额不足或服务被禁用
Access denied, insufficient balance or service disabled
429 Rate Limited
请求过于频繁,请降低请求频率
Too many requests, slow down your requests
500 Server Error
服务端内部错误,请稍后重试
Internal server error, please try again later

常见问题

FAQ

Q: 如何获取 API Key?
Q: How to get an API Key?
A: 注册并登录控制台后,在「API Keys」页面可以创建和管理您的 API Key。
A: After registering and logging into the console, create and manage your API Keys on the "API Keys" page.
Q: 支持流式输出(Streaming)吗?
Q: Does it support Streaming responses?
A: 是的,设置 stream: true 即可启用 SSE 流式返回。
A: Yes, set stream: true to enable SSE streaming responses.
Q: 如何查看用量和账单?
Q: How to view usage and billing?
A: 登录控制台后可在「用量统计」页面查看详细的 API 调用记录和费用明细。
A: After logging in, view detailed API usage and billing on the "Usage Statistics" page.
Q: 支持图片输入吗?
Q: Does it support image input?
A: 部分模型支持多模态输入,请在控制台查看具体模型的能力说明。
A: Some models support multimodal input. Check model capabilities in the console.

智能体接入指南

Agent Integration Guide

行业智能体提供Web界面和API两种使用方式,具体接入方式请咨询商务团队获取详细方案。

Industry agents provide both Web interface and API access. Contact our sales team for detailed integration plans.

1

选择行业智能体

Choose Industry Agent

Choose Industry Agent

根据业务需求选择合适的行业智能体,或联系商务定制开发

Select the right agent for your needs, or contact us for custom development

2

配置业务参数

Configure Parameters

Configure Parameters

根据企业实际情况配置行业参数、知识库和业务规则

Configure industry parameters, knowledge base and business rules

3

接入使用

Integrate & Use

Integrate & Use

通过Web界面直接使用,或通过API接入您的业务系统

Use via Web interface directly, or integrate with your system via API

💡 使用方式:💡 Access Methods:

Web界面 — 直接在浏览器中使用,适合快速体验和日常办公

API接入 — 通过标准REST API与您的业务系统集成,支持批量处理

Web Interface — Use directly in browser, perfect for quick testing and daily work

API Integration — Integrate with your systems via REST API, supports batch processing