📖 AI Token Hub API 文档

完整兼容 OpenAI API 格式，只需更换 base_url 和 api_key 即可接入全部中国主流大模型。

POST https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1/chat/completions

快速开始（30秒接入）

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-key-here",
    base_url="https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1"
)

response = client.chat.completions.create(
    model="deepseek-v4",  # 或 qwen3.6-flash, glm-4-flash 等
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "你好，请介绍一下自己"}
    ],
    temperature=0.7
)
print(response.choices[0].message.content)

curl

curl https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-key-here" \
  -d '{
    "model": "deepseek-v4",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Node.js

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'sk-your-key-here',
  baseURL: 'https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1'
});

const response = await client.chat.completions.create({
  model: 'deepseek-v4',
  messages: [
    { role: 'user', content: '你好！' }
  ]
});
console.log(response.choices[0].message.content);

LangChain

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="deepseek-v4",
    api_key="sk-your-key-here",
    base_url="https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1"
)

response = llm.invoke("你好，请介绍一下自己")
print(response.content)

LlamaIndex

from llama_index.llms.openai import OpenAI
from llama_index.core import Settings

Settings.llm = OpenAI(
    model="deepseek-v4",
    api_key="sk-your-key-here",
    api_base="https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1"
)

可用模型

模型 ID	提供商	输入价格 ($/1M tokens)	输出价格 ($/1M tokens)	特点
`deepseek-v4`	DeepSeek	$0.27	$1.10	旗舰模型，综合能力最强
`deepseek-v3`	DeepSeek	$0.27	$1.10	与V4性能相近
`deepseek-r1`	DeepSeek	$0.50	$2.00	推理增强，数学/逻辑
`deepseek-v4-flash`	DeepSeek	$0.001	$0.001	极速低价，高吞吐场景
`qwen3.6-flash`	Alibaba	$0.01	$0.03	中文优化，性价比高
`qwen3-flash`	Alibaba	$0.01	$0.03	快速响应
`qwen-plus`	Alibaba	$0.10	$0.10	均衡型
`qwen-max`	Alibaba	$0.50	$0.50	高质量输出
`glm-4-flash`	Zhipu	免费	免费	完全免费！
`glm-4-air`	Zhipu	$0.10	$0.10	平衡性能与成本
`kimi-k3`	Moonshot	$0.50	$2.00	长上下文，复杂推理

查询可用模型

GET /v1/models

curl https://eaf9553505eeb8f5-115-190-107-107.serveousercontent.com/v1/models \
  -H "Authorization: Bearer sk-your-key-here"

流式输出 (Streaming)

支持 SSE (Server-Sent Events) 流式输出，设置 stream: true：

response = client.chat.completions.create(
    model="deepseek-v4",
    messages=[{"role": "user", "content": "写一首诗"}],
    stream=True
)
for chunk in response:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

错误码说明

HTTP 状态码	含义	处理方式
401	API Key 无效或已过期	检查 API Key 是否正确
403	无权限访问该模型	确认套餐包含该模型
429	请求频率超限	降低请求频率或升级套餐
500	服务端错误	稍后重试或联系我们
503	模型暂时不可用	切换其他模型或稍后重试

💡 提示：所有请求和响应格式完全兼容 OpenAI API，您只需更改 base_url 和 api_key，无需修改任何业务代码。

免费试用

注册即可获得 50,000 tokens 免费额度，体验所有模型。

立即免费试用 →

← 返回首页 | 博客 | 成本计算器