Codex (OpenAI) 配置

配置 OpenAI Codex / OpenAI CLI 类工具，通过 QiyuanHub 统一调用模型。

准备工作

更省事的方式：用「CC Switch 统一配置」导入 Codex 应用类型；下方为手动环境变量写法。

Node.js 18+ 已安装
QiyuanHub 控制台已创建 API Key
本地已安装目标 Codex / OpenAI CLI 工具

在 CC Switch 中切换到 Codex — 若使用 CC Switch：顶部切换到 Codex（OpenAI 图标），再添加 / 导入供应商。

配置 OpenAI 兼容环境变量

1. macOS / Linux

export OPENAI_API_KEY="sk-opc-your-api-key"
export OPENAI_BASE_URL="https://api.qiyuanapi.cc/v1"

# 部分工具读取 AZURE_OPENAI_* 或自定义变量，请对照工具文档

2. Windows PowerShell

$env:OPENAI_API_KEY="sk-opc-your-api-key"
$env:OPENAI_BASE_URL="https://api.qiyuanapi.cc/v1"

3. 写入配置文件（持久化）

CC Switch / Codex 填写 QiyuanHub — 通过 CC Switch 配置 Codex 时：API 请求地址填 https://api.qiyuanapi.cc/v1；默认模型旁点下载按钮拉取模型列表。

# ~/.zshrc 或 ~/.bashrc
export OPENAI_API_KEY="sk-opc-your-api-key"
export OPENAI_BASE_URL="https://api.qiyuanapi.cc/v1"

Node.js SDK 示例

import OpenAI from "openai";

const client = new OpenAI({
 apiKey: process.env.OPENAI_API_KEY,
 baseURL: process.env.OPENAI_BASE_URL,
});

const completion = await client.chat.completions.create({
 model: "deepseek-v4.1",
 messages: [{ role: "user", content: "用 Python 写一个快速排序" }],
});

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

Python SDK 示例

from openai import OpenAI

client = OpenAI(
 api_key="sk-opc-your-api-key",
 base_url="https://api.qiyuanapi.cc/v1",
)

response = client.chat.completions.create(
 model="deepseek-v4.1",
 messages=[{"role": "user", "content": "你好"}],
)

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

常见问题

若工具仍请求 api.openai.com，检查是否有硬编码 Base URL
quota_exceeded 表示余额不足，请前往控制台钱包充值
model_not_allowed 表示当前账户无权访问该模型

import OpenAI from "openai"; const client = new OpenAI({ apiKey: process.env.OPENAI_API_KEY, baseURL: process.env.OPENAI_BASE_URL, }); const completion = await client.chat.completions.create({ model: "deepseek-v4.1", messages: [{ role: "user", content: "用 Python 写一个快速排序" }], }); console.log(completion.choices[0].message.content);

from openai import OpenAI client = OpenAI( api_key="sk-opc-your-api-key", base_url="https://api.qiyuanapi.cc/v1", ) response = client.chat.completions.create( model="deepseek-v4.1", messages=[{"role": "user", "content": "你好"}], ) print(response.choices[0].message.content)