Skip to content

README_en.md

Latest commit

 

History

History
276 lines (205 loc) · 9.18 KB

README_en.md

File metadata and controls

276 lines (205 loc) · 9.18 KB
Raw

Aix-DB

Aix-DB - LLM Data Assistant

An intelligent data analytics system powered by Large Language Models and RAG technology, enabling conversational data analysis (ChatBI) for rapid data extraction and visualization

Release Version GitHub Stars License Docker Pulls

简体中文 | English

🚀 Looking for Enterprise AI Solutions?

AiX-Bot

Our commercial product with powerful enterprise features:
Private Deployment · Custom Development · Dedicated Support · Multi-scenario AI Applications

👇 Click to Experience Now 👇

AI Chat Data Q&A Report Generation

💼 For business inquiries, please contact us via WeChat (note "Business Cooperation") | Contact Us

Aix-DB is built on the LangChain/LangGraph framework, combined with MCP Skills multi-agent collaboration architecture, enabling end-to-end transformation from natural language to data insights.

Core Capabilities: General Q&A · Data Q&A (Text2SQL) · Spreadsheet Q&A · Deep Research · Data Visualization · MCP Multi-Agent

Product Features: 📦 Ready to Use · 🔒 Secure & Controllable · 🔌 Easy Integration · 🎯 Increasingly Accurate

Demo Video

🎯 Skill Mode 💬 Standard Mode
2.7.1.2.mp4
default.mp4

System Architecture

System Architecture

Layered Architecture Design:

  • Frontend Layer: Modern web interface built with Vue 3 + TypeScript, integrated with ECharts and AntV visualization components
  • API Gateway Layer: High-performance async API service based on Sanic, providing RESTful interfaces and JWT authentication
  • Intelligent Service Layer: LLM services, Text2SQL Agent, RAG retrieval engine, MCP multi-agent collaboration
  • Data Storage Layer: Support for multiple database types including relational databases, vector databases, graph databases, and file storage

Supported Data Sources

Data Q&A Workflow

Step Module Description
1 User Input User asks data query questions in natural language
2 LLM Intent Understanding LLM parses question intent, extracts key entities and query conditions
3 RAG Knowledge Retrieval Embedding + BM25 hybrid retrieval, combined with Neo4j graph to obtain relevant table structures and business knowledge
4 SQL Generation Text2SQL engine generates SQL statements with syntax validation and optimization
5 Database Execution Execute SQL on target data source, supporting 8+ database types
6 Visualization Automatically generate ECharts/AntV charts to present analysis results

Quick Start

Deploy with Docker (Recommended)

docker run -d \
  --name aix-db \
  --restart unless-stopped \
  -e TZ=Asia/Shanghai \
  -e SERVER_HOST=0.0.0.0 \
  -e SERVER_PORT=8088 \
  -e SERVER_WORKERS=2 \
  -e LANGFUSE_TRACING_ENABLED=false \
  -e LANGFUSE_SECRET_KEY= \
  -e LANGFUSE_PUBLIC_KEY= \
  -e LANGFUSE_BASE_URL= \
  -p 18080:80 \
  -p 18088:8088 \
  -p 15432:5432 \
  -p 9000:9000 \
  -p 9001:9001 \
  -v ./volume/pg_data:/var/lib/postgresql/data \
  -v ./volume/minio/data:/data \
  -v ./volume/logs/supervisor:/var/log/supervisor \
  -v ./volume/logs/nginx:/var/log/nginx \
  -v ./volume/logs/aix-db:/var/log/aix-db \
  -v ./volume/logs/minio:/var/log/minio \
  -v ./volume/logs/postgresql:/var/log/postgresql \
  --add-host host.docker.internal:host-gateway \
  crpi-7xkxsdc0iki61l0q.cn-hangzhou.personal.cr.aliyuncs.com/apconw/aix-db:1.2.3

Note: To enable Langfuse full-chain tracing, set LANGFUSE_TRACING_ENABLED=true and configure the corresponding keys and URL.

Deploy with Docker Compose

git clone https://github.com/apconw/Aix-DB.git
cd Aix-DB/docker
cp .env.template .env  # Copy env template, modify as needed
docker-compose up -d

Access the System

Web Management Interface

PostgreSQL Database

  • Connection: localhost:15432
  • Database: aix_db
  • Username: aix_db
  • Password: 1

Local Development

① Clone the Repository

git clone https://github.com/apconw/Aix-DB.git
cd Aix-DB

② Start Middleware Dependencies (PostgreSQL, MinIO, etc.)

cd docker
docker-compose up -d

③ Configure Environment Variables

Edit .env.dev in the project root to set database connection, MinIO address, etc. (default config works out of the box)

④ Install Python Dependencies (requires Python 3.11)

# Option 1: pip
pip install -r requirements.txt

# Option 2: uv (recommended, faster)
uv venv --python 3.11
source .venv/bin/activate
uv sync

⑤ Start Backend Service

python serv.py

⑥ Start Frontend Dev Server (in another terminal)

cd web
npm install
npm run dev

Tech Stack

Backend: Sanic · SQLAlchemy · LangChain/LangGraph · Neo4j · FAISS/Chroma · MinIO

Frontend: Vue 3 · TypeScript · Vite 5 · Naive UI · ECharts · AntV

AI Models: OpenAI · Anthropic · DeepSeek · Qwen · Ollama

Documentation

Contributing

We welcome Issues and Pull Requests!

  1. Fork this repository
  2. Create a feature branch (git checkout -b feature/AmazingFeature)
  3. Commit your changes (git commit -m 'Add some AmazingFeature')
  4. Push to the branch (git push origin feature/AmazingFeature)
  5. Open a Pull Request

Contact Us

If you have any questions, feel free to reach out:

Star History

Star History Chart

License

This project is licensed under the Apache License 2.0.