AI Chat API

A FastAPI backend for an AI chat application with support for multiple LLM models and conversation history.

Features

• Multiple LLM model support (deepseek-r1, qwen)
• Conversation history management
• Redis-based caching for chat history
• MySQL database for structured data
• Authentication and authorization
• RESTful API endpoints

Prerequisites

• Python 3.8+
• MySQL 8.0+
• Redis
• Ollama (for LLM models)

Installation

1. Clone the repository:

git clone <repository-url>
cd ai-chat-api

2. Create a virtual environment and activate it:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file with the following variables:

SECRET_KEY=your-secret-key
MYSQL_HOST=localhost
MYSQL_PORT=3306
MYSQL_USER=root
MYSQL_PASSWORD=your-password
MYSQL_DB=chat_db
REDIS_HOST=localhost
REDIS_PORT=6379
REDIS_DB=0

5. Initialize the database:

python init_db.py

Running the Application

Start the FastAPI server:

uvicorn main:app --reload

The API will be available at http://localhost:8000

API Documentation

Once the server is running, you can access:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

API Endpoints

Chat

• POST /api/v1/chat/conversations - Create a new conversation
• GET /api/v1/chat/conversations - List user's conversations
• POST /api/v1/chat/conversations/{conversation_id}/messages - Send a message
• GET /api/v1/chat/conversations/{conversation_id}/messages - Get conversation messages

Project Structure

.
├── main.py
├── requirements.txt
├── README.md
├── core/
│   └── config.py
├── models/
│   ├── base.py
│   └── chat.py
├── routers/
│   └── chat.py
├── schemas/
│   └── chat.py
├── services/
│   └── chat.py
└── dependencies/
    └── auth.py

项目架构

1. 采用分层架构：
   • routers/ (API端点)
   • models/ (Pydantic和SQLAlchemy模型)
   • schemas/ (数据验证模型)
   • services/ (业务逻辑)
   • utils/ (工具函数)
   • dependencies/ (依赖注入)
   • core/ (配置和安全)