README_en.md

Getting Started

LangChain4j-AIDeepin is an AI-based productivity enhancement tool.

It can be used to assist enterprises/teams in technical research and development, product design, HR/finance/IT information consulting, system/product consulting, customer service support, etc.

🌟 If this project is helpful to you, please give it a star 🌟

System Composition and Documentation

中文文档 | English

AIDEEPIN

      |__ Server (langchain4j-aideepin)

      |__ User Web (langchain4j-aideepin-web)

      |__ Admin Web (langchain4j-aideepin-admin)

👉Detailed Documentation

Backend source repository: github or gitee

Frontend projects:

• User Web: langchain4j-aideepin-web
  • github
  • gitee
• Admin Web: langchain4j-aideepin-admin
  • github
  • gitee

Demo URL

http://www.aideepin.com

Features

• Registration & Login
• Multi-session (multi-role)
• Image generation (text-to-image, image editing, image-to-image)
• Prompts
• Quota control
• Knowledge base based on large models (RAG)
  • Vector search
  • Graph search
• Search based on large models (RAG)
• Free switching between multiple models
• Free switching between multiple search engines

Integrated Models:

• ChatGPT 3.5
• 通义千问
• 文心一言
• Ollama
• DALL-E 2
• DALL-E 3

Integrated Search Engines

Google

Bing (TODO)

Baidu (TODO)

Tech Stack

This repository is for the backend service

Tech stack:

• jdk17
• springboot3.0.5
• langchain4j (Java version of LangChain)
• Postgresql
  • pgvector extension: https://github.com/pgvector/pgvector
  • Apache AGE extension: https://github.com/apache/age

Frontend tech stack:

• vue3
• vite
• typescript
• pnpm
• pinia
• naiveui

How to Deploy

Initialization

a. Initialize the database

• Create the database aideepin

• Execute docs/create.sql

• Configure models (at least set one) or use the admin web to configure on the interface

  • Configure AI platforms

    -- OpenAI secretKey
    update adi_sys_config set value = '{"secret_key":"my_openai_secret_key"}' where name = 'openai_setting';
    
    -- Dashscope API key
    update adi_sys_config set value = '{"api_key":"my_dashcope_api_key"}' where name = 'dashscope_setting';
    
    -- Qianfan API key and secret key
    update adi_sys_config set value = '{"api_key":"my_qianfan_api_key","secret_key":"my_qianfan_secret_key"}' where name = 'qianfan_setting';
    
    -- Ollama configuration
    update adi_sys_config set value = '{"base_url":"my_ollama_base_url"}' where name = 'ollama_setting';
    

  • Enable AI platform models or add new models

    -- Enable model
    update adi_ai_model set is_enable = true where name = 'gpt-3.5-turbo';
    update adi_ai_model set is_enable = true where name = 'dall-e-2';
    update adi_ai_model set is_enable = true where name = 'qwen-turbo';
    update adi_ai_model set is_enable = true where name = 'ernie_speed';
    update adi_ai_model set is_enable = true where name = 'tinydolphin';
    
    -- Add new model
    INSERT INTO adi_ai_model (name, type, platform, is_enable) VALUES ('vicuna', 'text', 'ollama', true);
    

• Fill in the search engine configuration

  • Google configuration

    update adi_sys_config set value = '{"url":"https://www.googleapis.com/customsearch/v1","key":"my key from cloud.google.com","cx":"my cx from programmablesearchengine.google.com"}' where name = 'google_setting';
    

b. Modify the configuration file

• postgresql: application-[dev|prod].xml in spring.datasource
• redis: application-[dev|prod].xml in spring.data.redis
• mail: application.xml in spring.mail

Build and Run

• Enter the project

  cd langchain4j-aideepin
  

• Package:

  mvn clean package -Dmaven.test.skip=true
  

• Run

  • Start with jar:

  cd adi-bootstrap/target
  nohup java -jar -Xms768m -Xmx1024m -XX:+HeapDumpOnOutOfMemoryError adi-bootstrap-0.0.1-SNAPSHOT.jar --spring.profiles.active=[dev|prod] dev/null 2>&1 &
  

  • Start with docker

  cd adi-bootstrap
  docker build . -t aideepin:0.0.1
  docker run -d \
    --name=aideepin \
    -e APP_PROFILE=[dev|prod] \
    -v="/data/aideepin/logs:/data/logs" \
    aideepin:0.0.1
  

TODO:

• Advanced RAG
  • Query compression √
  • Query routing
  • Re-rank: support local rerank model
• AI Chat
  • Multi-role √
  • Preset general roles (created in admin backend) √
• Image Models:
  • DALL-E 2 & DALL-E 3 √
  • Chat view √
  • Gallery view √
  • Open/collect images √
  • Image comments √
• Knowledge Base:
  • Vector √
  • Knowledge graph √
  • Set the number of document recalls
    • Auto-adjust (based on LLM context window size) √
    • Manual adjustment √
  • Set the minimum score for document recall √
  • Set the overlap number when chunking √
  • Set temperature when requesting model √
  • Strict mode and non-strict mode √
  • Answer source √
  • Support fetching online documents
  • FAQ
  • Comments
• Multi-modal support
  • Image √
  • Audio
  • Video
• Tools
  • FAQ extraction
  • Document conversation
  • Translation
  • Image processing
    • Image editing
    • Image matting
    • Image expansion
    • Image synthesis
    • Background switching
• Search Engines
  • Google √
  • Bing
  • Baidu
• Quota statistics and control
  • Free quota statistics and limits
  • Paid quota statistics and limits
  • Total quota statistics
• Open API

Screenshots

AI Chat:

AI Drawing:

Knowledge Base:

Vectorization:

Knowledge Graph:

Quota Statistics: