快速开始

快速开始

选择你的应用使用的协议。只需要配置该协议对应的 Base URL。

准备信息

项目值
API KeyJOY_TOKEN_API_KEY=sk-xxx
模型auto（不可配置）

选择协议接口

Base URL：https://api.joytokens.ai/openai/v1

export JOY_TOKEN_API_KEY="sk-xxx"
export JOY_TOKEN_OPENAI_BASE_URL="https://api.joytokens.ai/openai/v1"

使用 cURL 调用 Responses API

curl -sS -X POST "$JOY_TOKEN_OPENAI_BASE_URL/responses" \
  -H "Authorization: Bearer $JOY_TOKEN_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-Request-ID: quickstart-responses-001" \
  -d '{
    "model": "auto",
    "input": "只回复：pong",
    "max_output_tokens": 1024,
    "stream": false
  }'

使用 OpenAI SDK 调用 Responses API

import OpenAI from "openai";
const client = new OpenAI({
  apiKey: process.env.JOY_TOKEN_API_KEY,
  baseURL: process.env.JOY_TOKEN_OPENAI_BASE_URL,
});
const response = await client.responses.create({
  model: "auto",
  input: "只回复：pong",
  max_output_tokens: 1024,
});
console.log(response.output_text);

使用 cURL 调用 Chat Completions

curl -sS -X POST "$JOY_TOKEN_OPENAI_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $JOY_TOKEN_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-Request-ID: quickstart-001" \
  -d '{
    "model": "auto",
    "messages": [
      { "role": "user", "content": "只回复：pong" }
    ],
    "temperature": 0.7,
    "max_tokens": 1024,
    "stream": false
  }'

使用 OpenAI SDK 调用 Chat Completions

import OpenAI from "openai";
const client = new OpenAI({
  apiKey: process.env.JOY_TOKEN_API_KEY,
  baseURL: process.env.JOY_TOKEN_OPENAI_BASE_URL,
});
const response = await client.chat.completions.create({
  model: "auto",
  messages: [{ role: "user", content: "只回复：pong" }],
  max_tokens: 1024,
});
console.log(response.choices[0]?.message?.content);

检查接入结果

位置应看到
程序输出响应包含 pong
响应 metadatametadata.model 和 metadata.billing
JoyToken ConsoleLogs 和 Usage 归因到当前 API Key

API Key 应保存在服务端环境变量、密钥管理服务或 CI/CD Secrets 中，不要放进浏览器、移动端、公开仓库或日志。