### Local Installation and Running Qwen-Proxy

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

This guide details the steps for setting up Qwen-Proxy on a local machine. It includes cloning the repository, installing Node.js dependencies using npm, copying and configuring the `.env` file, and starting the service using `npm start` for intelligent startup or `npm run dev` for development mode.

```bash
# 克隆项目
git clone https://github.com/Rfym21/Qwen2API.git
cd Qwen2API

# 安装依赖
npm install

# 配置环境变量
cp .env.example .env
# 编辑 .env 文件

# 智能启动 (推荐 - 自动判断单进程/多进程)
npm start

# 开发模式
npm run dev

```

--------------------------------

### Docker Compose for Qwen-Proxy Deployment

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

This section provides instructions for deploying Qwen-Proxy using Docker Compose. It involves downloading a `docker-compose.yml` file and then using `docker compose` commands to pull images and start the service. This method is suitable for more complex deployments or for managing multiple related services.

```bash
# 下载配置文件
curl -o docker-compose.yml https://raw.githubusercontent.com/Rfym21/Qwen2API/refs/heads/main/docker-compose.yml

# 启动服务
docker compose pull && docker compose up -d

```

--------------------------------

### API Authentication Example (Bash)

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Demonstrates how to authenticate API requests using both administrator and regular API keys via curl. The Authorization header is used to pass the API key.

```bash
# 使用管理员密钥
curl -H "Authorization: Bearer sk-admin123" http://localhost:3000/v1/models

# 使用普通密钥
curl -H "Authorization: Bearer sk-user456" http://localhost:3000/v1/chat/completions
```

--------------------------------

### GET /v1/models

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Retrieves a list of all available AI models supported by the API.

```APIDOC
## 🔍 获取模型列表

获取所有可用的 AI 模型列表。

```http
GET /v1/models
Authorization: Bearer sk-your-api-key
```

```http
GET /models (免认证)
```

**响应示例:**
```json
{
  "object": "list",
  "data": [
    {
      "id": "qwen-max-latest",
      "object": "model",
      "created": 1677610602,
      "owned_by": "qwen"
    }
  ]
}
```
```

--------------------------------

### Get Model List API Request (HTTP)

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Shows how to make a GET request to retrieve a list of available AI models. Authentication is required for some endpoints, while others are publicly accessible.

```http
GET /v1/models
Authorization: Bearer sk-your-api-key
```

```http
GET /models (免认证)
```

--------------------------------

### Chat Completion Request Body (JSON)

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Provides an example JSON payload for the chat completions API. It specifies the model to use, the conversation history (messages), and generation parameters.

```json
{
  "model": "qwen-max-latest",
  "messages": [
    {
      "role": "system",
      "content": "你是一个有用的助手。"
    },
    {
      "role": "user",
      "content": "你好，请介绍一下自己。"
    }
  ],
  "stream": false,
  "temperature": 0.7,
  "max_tokens": 2000
}
```

--------------------------------

### Image Generation Request Body (JSON)

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

An example JSON payload for image generation. It includes the model, user prompt, and optional parameters like 'size' for image dimensions and 'stream' for response format.

```json
{
  "model": "qwen-max-latest-image",
  "messages": [
    {
      "role": "user",
      "content": "画一只在花园里玩耍的小猫咪，卡通风格"
    }
  ],
  "size": "1:1",
  "stream": false
}
```

--------------------------------

### Docker Run Command for Qwen-Proxy

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

This command demonstrates how to run the Qwen-Proxy service using Docker. It maps ports, sets necessary environment variables for API keys, data saving, and cache modes, and mounts a volume for cache persistence. This is a direct way to deploy the service in a containerized environment.

```bash
docker run -d \
  -p 3000:3000 \
  -e API_KEY=sk-admin123,sk-user456,sk-user789 \
  -e DATA_SAVE_MODE=none \
  -e CACHE_MODE=file \
  -e ACCOUNTS= \
  -v ./caches:/app/caches \
  --name qwen2api \
  rfym21/qwen2api:latest

```

--------------------------------

### Configure Environment Variables for Qwen-Proxy

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

This snippet shows the essential environment variables required to configure the Qwen-Proxy service. These variables control network settings, security, performance optimizations, and data persistence. Ensure these are set in a `.env` file for local deployments or passed as environment variables in containerized environments.

```bash
# 🌐 服务配置
LISTEN_ADDRESS=localhost       # 监听地址
SERVICE_PORT=3000             # 服务端口

# 🔐 安全配置
API_KEY=sk-123456,sk-456789   # API 密钥 (必填，支持多密钥)
ACCOUNTS=                     # 账户配置 (格式: user1:pass1,user2:pass2)

# 🚀 PM2 多进程配置
PM2_INSTANCES=1               # PM2进程数量 (1/数字/max)
PM2_MAX_MEMORY=1G             # PM2内存限制 (100M/1G/2G等)
                              # 注意: PM2集群模式下所有进程共用同一个端口

# 🔍 功能配置
SEARCH_INFO_MODE=table        # 搜索信息展示模式 (table/text)
OUTPUT_THINK=true             # 是否输出思考过程 (true/false)
SIMPLE_MODEL_MAP=false        # 简化模型映射 (true/false)

# 🗄️ 数据存储
DATA_SAVE_MODE=none           # 数据保存模式 (none/file/redis)
REDIS_URL=                    # Redis 连接地址 (可选)

# 📸 缓存配置
CACHE_MODE=default            # 图片缓存模式 (default/file)

```

--------------------------------

### POST /v1/chat/completions (Image Generation/Editing)

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Enables text-to-image generation using `-image` models and image editing using `-image-edit` models. Allows specifying image size and stream options.

```APIDOC
## 🎨 图像生成/编辑

使用 `-image` 模型启用文本到图像生成功能。
使用 `-image-edit` 模型启用图像修改功能。
当使用``-image` 时你可以通过在请求体中添加 `size` 参数或在消息内容中包含特定关键词 `1:1`, `4:3`, `3:4`, `16:9`, `9:16` 来控制图片尺寸。

```http
POST /v1/chat/completions
Content-Type: application/json
Authorization: Bearer sk-your-api-key
```

**请求体:**
```json
{
  "model": "qwen-max-latest-image",
  "messages": [
    {
      "role": "user",
      "content": "画一只在花园里玩耍的小猫咪，卡通风格"
    }
  ],
  "size": "1:1",
  "stream": false
}
```

**支持的参数:**
- `size`: 图片尺寸，支持 `"1:1"`、`"4:3"`、`"3:4"`、`"16:9"`、`"9:16"`
- `stream`: 支持流式和非流式响应

**响应示例:**
```json
{
  "created": 1677652288,
  "model": "qwen-max-latest",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "![image](https://example.com/generated-image.jpg)"
      },
      "finish_reason": "stop"
    }
  ]
}
```
```

--------------------------------

### API Authentication

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Explains the API's multi-key authentication mechanism and how to include the API key in requests.

```APIDOC
## 🔐 API 认证说明

本API支持多密钥认证机制，所有API请求都需要在请求头中包含有效的API密钥：

```http
Authorization: Bearer sk-your-api-key
```

**支持的密钥类型:**
- **管理员密钥**: 第一个配置的API_KEY，拥有完整权限
- **普通密钥**: 其他配置的API_KEY，仅可调用API接口

**认证示例:**
```bash
# 使用管理员密钥
curl -H "Authorization: Bearer sk-admin123" http://localhost:3000/v1/models

# 使用普通密钥
curl -H "Authorization: Bearer sk-user456" http://localhost:3000/v1/chat/completions
```
```

--------------------------------

### Chat Completion API Request (HTTP)

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Defines the structure for making a POST request to the chat completions endpoint. This includes the model, messages, and optional parameters like temperature and max_tokens.

```http
POST /v1/chat/completions
Content-Type: application/json
Authorization: Bearer sk-your-api-key
```

--------------------------------

### POST /v1/chat/completions

Source: https://github.com/rfym21/qwen2api/blob/main/README.md

Allows users to send chat messages and receive AI-generated responses. Supports various parameters for controlling the conversation.

```APIDOC
## 💬 聊天对话

发送聊天消息并获取 AI 回复。

```http
POST /v1/chat/completions
Content-Type: application/json
Authorization: Bearer sk-your-api-key
```

**请求体:**
```json
{
  "model": "qwen-max-latest",
  "messages": [
    {
      "role": "system",
      "content": "你是一个有用的助手。"
    },
    {
      "role": "user",
      "content": "你好，请介绍一下自己。"
    }
  ],
  "stream": false,
  "temperature": 0.7,
  "max_tokens": 2000
}
```

**响应示例:**
```json
{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "qwen-max-latest",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "你好！我是一个AI助手..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 20,
    "completion_tokens": 50,
    "total_tokens": 70
  }
}
```
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.