GitHub - sjkncs/infortune-ai-desktop: In Fortune AI - Professional AI-powered stock analysis desktop application with multi-model support, real-time market data (BaoStock/East Money), strategy backtesting, and portfolio management

Note

In Fortune AI is a professional AI-powered stock analysis platform featuring a desktop client (Electron) and a web app (Vue 3). It integrates 20+ AI models (Gemini, GPT, Grok), real-time Chinese A-share market data via BaoStock & East Money, strategy backtesting, and portfolio management.

Tip

Based on @QoneFeng's open-source project InFortune.ai, extensively enhanced with a full Electron desktop app, multi-model AI support, and real-time data integration.

🚀 What's New in v1.0.0

🤖 Multi-Model AI Engine 20+ AI models with in-chat model selector Gemini 2.5/3 Pro & Flash (dialogue, search, tool-calling) GPT-4o, GPT-5, GPT-5.2, Grok-4 Image generation (Gemini Flash/Pro) Model selection persisted per session	📈 Real-Time Market Data 3-tier data fallback: Backend → BaoStock → East Money Shanghai Composite, Shenzhen Component, ChiNext, CSI 300 BaoStock direct K-line fetching (port 8001) East Money free API for real-time index quotes Auto-refresh with configurable intervals
💬 Enhanced Chat Interface File upload (images, PDF, Excel, CSV, Markdown) Voice input via Web Speech API Chat history with timestamps (like Doubao/Qwen) Categorized model dropdown with descriptions	🖥️ Desktop App Improvements Fixed Windows keyboard input/IME in Electron Updated CSP for multi-source data access Role-based Electron menus for full compatibility Detached DevTools to prevent focus stealing

🤖 Multi-Model AI Engine

In Fortune AI integrates 20+ AI models across 5 categories, selectable directly from the chat interface:

┌──────────────────────────────────────────────────────────┐
│  🤖 Model Selector                              ▼       │
├──────────────────────────────────────────────────────────┤
│  💬 DIALOGUE    │ Gemini 3 Flash / 2.5 Pro / 3 Pro      │
│  🔍 SEARCH      │ Gemini 3 Pro Search / 2.5 Flash Search│
│  🔧 TOOL-CALL   │ Gemini 3 Pro (Tool) / 2.5 Flash (Tool)│
│  🤖 GPT / GROK  │ GPT-5 / GPT-4o / GPT-5.2 / Grok-4    │
│  🎨 IMAGE GEN   │ Gemini Flash 生图 / Gemini Pro 生图    │
└──────────────────────────────────────────────────────────┘

📋 Complete Model Catalog (click to expand)

Category	Model ID	Name	Description
💬 Dialogue	`gemini-3-flash`	Gemini 3 Flash	Best cost-performance, half the price of 2.5 Pro
💬 Dialogue	`gemini-2.5-pro`	Gemini 2.5 Pro	1M context, built-in chain-of-thought + search
💬 Dialogue	`gemini-3-pro`	Gemini 3 Pro	Latest Pro model, slightly higher cost
💬 Dialogue	`gemini-2.5-flash`	Gemini 2.5 Flash	Fast response model
🔍 Search	`gemini-3-pro-search`	Gemini 3 Pro Search	Real-time web search built-in
🔍 Search	`gemini-2.5-pro-search`	Gemini 2.5 Pro Search	Real-time web search built-in
🔍 Search	`gemini-2.5-flash-search`	Gemini 2.5 Flash Search	Fast search model (default)
🔍 Search	`gemini-3-flash-search`	Gemini 3 Flash Search	Fast search model
🔧 Tool	`gemini-3-pro-no`	Gemini 3 Pro (Tool)	Search disabled, custom tools supported
🔧 Tool	`gemini-2.5-pro-no`	Gemini 2.5 Pro (Tool)	Search disabled, custom tools supported
🔧 Tool	`gemini-2.5-flash-no`	Gemini 2.5 Flash (Tool)	Search disabled, custom tools supported
🔧 Tool	`gemini-3-flash-no`	Gemini 3 Flash (Tool)	Search disabled, custom tools supported
🤖 GPT	`gpt-5`	GPT-5	OpenAI's latest flagship
🤖 GPT	`gpt-4o`	GPT-4o	OpenAI multimodal model
🤖 GPT	`gpt-5.2`	GPT-5.2	OpenAI advanced version
🤖 GPT	`grok-4`	Grok-4	xAI Grok model
🎨 Image	`gemini-2.5-flash-image-preview`	Gemini Flash Image Gen	~$0.01/image, English prompts recommended
🎨 Image	`gemini-3-pro-image-preview`	Gemini Pro Image Gen	High-quality, ~$0.04/image

Important

The model selector remembers your last choice via localStorage. The selected model is sent to the backend /api/chat endpoint as the model parameter. All models are served through the HiAPI unified gateway (https://hiapi.online/v1).

📈 Real-Time Market Data

Market data uses a 3-tier fallback architecture for maximum reliability:

graph LR
    A[Desktop App] --> B{Source 1: Backend API}
    B -->|Success| Z[Display Data]
    B -->|Fail| C{Source 2: BaoStock Direct}
    C -->|Success| Z
    C -->|Fail| D{Source 3: East Money API}
    D -->|Success| Z
    D -->|Fail| E[All Sources Unavailable]
    
    style A fill:#1f6feb,color:#fff
    style Z fill:#238636,color:#fff
    style E fill:#da3633,color:#fff

Source	URL	Data Type	Latency
Backend Proxy	`localhost:3001/api/market/overview`	Aggregated indices	~200ms
BaoStock Direct	`127.0.0.1:8001/api/kline/history`	Daily K-line (OHLCV)	~500ms
East Money	`push2.eastmoney.com`	Real-time quotes	~300ms

📊 Supported Indices

Index	Code	BaoStock Code	East Money SecID
Shanghai Composite (上证指数)	`000001`	`sh.000001`	`1.000001`
Shenzhen Component (深证成指)	`399001`	`sz.399001`	`0.399001`
ChiNext (创业板指)	`399006`	`sz.399006`	`0.399006`
CSI 300 (沪深300)	`000300`	`sh.000300`	`1.000300`

🖥️ Desktop Application

10 View Modules

#	Module	View ID	Description
1	🤖 Smart Analysis	`chat`	AI-powered chat with model selector, file upload, voice input
2	📈 Market Overview	`market`	Real-time indices with multi-source data fallback
3	📊 Stock Analysis	`stock`	Individual stock search and fundamental/technical analysis
4	⭐ Watchlist	`zixuan`	Custom stock watchlist with real-time monitoring
5	💼 Portfolio	`portfolio`	Holdings management with P&L tracking
6	📉 Index Analysis	`zhishu`	Index valuation (PE/PB percentile analysis)
7	📦 ETF Market	`etf`	ETF listings with search, sort by volume/change/amount
8	🧪 Strategy Backtest	`strategy`	MA/MACD/KDJ/RSI/Bollinger strategies with equity curves
9	🔀 Mode Switch	`mode`	Toggle between Local and Web modes
10	⚙️ Settings	`settings`	8 setting panels: General, Appearance, Language, Data, etc.

Chat Interface Features

┌─────────────────────────────────────────────────────────────┐
│  📜 Chat History (sidebar)          │  Welcome / Messages   │
│  ─────────────────────              │                       │
│  💬 贵州茅台分析  14:30             │  [Model: Gemini 2.5]  │
│  💬 市场行情概览  13:15             │  ┌─────────────────┐  │
│  💬 技术指标讨论  11:20             │  │ 🤖 Model Selector│  │
│                                     │  └─────────────────┘  │
│                                     │  ┌─────────────────┐  │
│                                     │  │📎 [text] 🎤 ➤  │  │
│                                     │  └─────────────────┘  │
│                                     │  [实时数据][技术][新闻]│
└─────────────────────────────────────────────────────────────┘

Model Selector — categorized dropdown above input, persistent selection
File Upload — images, PDF, Word, Excel, CSV, Markdown
Voice Input — Web Speech API with recording indicator
Chat History — timestamped conversations saved to localStorage

🏗️ Architecture

                          ┌─────────────────────┐
                          │   Electron Desktop   │
                          │    (main.js + UI)    │
                          └──────────┬──────────┘
                                     │
              ┌──────────────────────┼──────────────────────┐
              │                      │                      │
     ┌────────▼────────┐   ┌────────▼────────┐   ┌────────▼────────┐
     │  Vue 3 Web App  │   │  Express Backend │   │  Admin Panel    │
     │  :5173           │   │  :3001           │   │  :5174          │
     └─────────────────┘   └────────┬─────────┘   └─────────────────┘
                                    │
                    ┌───────────────┼───────────────┐
                    │               │               │
           ┌────────▼──────┐ ┌─────▼──────┐ ┌──────▼───────┐
           │ HiAPI Gateway │ │  BaoStock  │ │  East Money  │
           │ (AI Models)   │ │  :8001     │ │   (Free API) │
           └───────────────┘ └────────────┘ └──────────────┘
                    │
        ┌───────────┼───────────┐
        │           │           │
   ┌────▼───┐ ┌────▼───┐ ┌────▼───┐
   │ Gemini │ │  GPT   │ │  Grok  │
   │ 2.5/3  │ │ 4o/5   │ │   4    │
   └────────┘ └────────┘ └────────┘

Service Ports

Service	Port	Technology	Purpose
Desktop App	—	Electron 28	Native desktop client
Web Frontend	`5173`	Vue 3 + Vite + TailwindCSS	Web interface
Admin Panel	`5174`	Vue 3 + Vite	Data management
API Server	`3001`	Express + LowDB + WebSocket	REST API + real-time push
AKShare Service	`8000`	FastAPI + AKShare	Multi-market stock data
BaoStock Service	`8001`	FastAPI + BaoStock	A-share financial data
ML Service	`8002`	Python	Stock prediction inference

⚡ Quick Start

Prerequisites

Node.js >= 18.0.0  |  Python >= 3.11  |  npm >= 9.0.0

1. Clone & Install

git clone https://github.com/sjkncs/infortune-ai-desktop.git
cd infortune-ai-desktop

# Desktop app
cd desktop-app && npm install

# Backend server
cd ../website && npm install

# BaoStock service (optional)
cd ../baostock && pip install -r requirements.txt

2. Configure

# Copy and edit .env
cp .env.example .env

AI_API_KEY=your_hiapi_key_here
AI_API_BASE_URL=https://hiapi.online/v1
AI_MODEL=gemini-2.5-flash-search
BAOSTOCK_API_URL=http://127.0.0.1:8001

3. Run

Terminal	Command	Service
1	`cd website && npm run server`	Backend API :3001
2	`cd baostock && python app.py`	BaoStock :8001
3	`cd desktop-app && npm start`	Desktop App

4. Build for Distribution

cd desktop-app
npm run build:win     # Windows (portable + zip)
# Output: dist/InFortuneAI-1.0.0-win-x64.exe

📡 API Reference

🔵 Core APIs — Node.js Backend (:3001)

Method	Endpoint	Description
`POST`	`/api/chat`	AI chat (accepts `model` param from frontend)
`GET`	`/api/market/overview`	Market indices overview
`GET`	`/api/search?keyword=`	Stock search (fuzzy match)
`GET`	`/api/stocks`	Stock list with real-time prices + MA
`GET`	`/api/portfolio`	Portfolio holdings with prices
`POST`	`/api/save-portfolio`	Save portfolio data
`GET`	`/api/stock/:code`	Stock detail (orderbook, valuation)
`GET`	`/api/stock/:code/kline`	K-line data (multi-period)
`GET`	`/api/stock/:code/realtime`	Single stock real-time quote
`GET`	`/api/indices/realtime?codes=`	Batch index real-time quotes
`GET`	`/api/etf`	ETF list via BaoStock
`GET`	`/api/valuation`	Index valuation (PE/PB)
`POST`	`/api/backtest`	Strategy backtesting
`POST`	`/api/ml/predict`	ML stock prediction
`GET`	`/api/health`	Health check

🟢 BaoStock APIs (:8001)

Endpoint	Description
`/api/stock/list`	A-share stock listing
`/api/etf/list`	ETF fund listing
`/api/kline/history?code=&start_date=&end_date=&frequency=`	Historical K-line (daily/weekly/monthly)
`/api/finance/profit/{code}`	Profitability (ROE, net margin, gross margin)
`/api/finance/growth/{code}`	Growth metrics
`/api/finance/balance/{code}`	Balance sheet ratios
`/api/finance/cash-flow/{code}`	Cash flow statement
`/api/finance/dupont/{code}`	DuPont analysis
`/api/macro/deposit-rate`	Deposit interest rates
`/api/macro/money-supply`	Money supply (M0/M1/M2)

🟡 AKShare APIs (:8000)

Endpoint	Description
`/api/a/stock/realtime/{symbol}`	A-share real-time quote
`/api/a/stock/history/{symbol}`	A-share historical K-line
`/api/a/stock/list`	A-share stock list
`/api/hk/stock/*`	Hong Kong stocks
`/api/us/stock/*`	US stocks
`/api/etf/*`	ETF data
`/api/index/*`	Index data

📁 Project Structure

📂 Click to expand full directory tree

infortune-ai-desktop/
├── 📁 desktop-app/              # Electron Desktop Application
│   ├── main.js                  #   Main process (window, menu, tray)
│   ├── preload.js               #   Preload script (context bridge)
│   ├── index.html               #   Main window (10 view containers)
│   ├── package.json             #   Electron + electron-builder config
│   ├── 📁 scripts/              #   Frontend modules
│   │   ├── app.js               #     App init, routing, model selector
│   │   ├── api.js               #     AI API calls (multi-model)
│   │   ├── api-config.js        #     MODEL_CATALOG + provider configs
│   │   ├── api-client.js        #     HTTP client with retry/stream
│   │   ├── chat.js              #     Chat UI logic
│   │   ├── market.js            #     Market data (3-source fallback)
│   │   ├── stock.js             #     Stock analysis module
│   │   ├── portfolio.js         #     Portfolio management
│   │   ├── zixuan.js            #     Watchlist module
│   │   ├── zhishu.js            #     Index valuation
│   │   ├── etf.js               #     ETF market module
│   │   ├── strategy.js          #     Strategy backtesting
│   │   ├── settings.js          #     Settings panel logic
│   │   └── i18n.js              #     Internationalization
│   └── 📁 styles/               #   CSS modules (CSS custom properties)
│       ├── main.css, chat.css, views.css, strategy.css,
│       ├── stock.css, portfolio.css, market.css,
│       ├── mode.css, settings.css
│
├── 📁 website/                  # Vue 3 Web Application
│   ├── 📁 src/
│   │   ├── 📁 views/            #   Home, Zixuan, Chicang, Zhishu,
│   │   │                        #   ETF, StockDetail, NotFound
│   │   ├── 📁 router/           #   Vue Router (7 routes)
│   │   ├── 📁 composables/      #   useStockApi, useAiAnalysis, etc.
│   │   └── 📁 components/       #   AppHeader, AppFooter
│   └── 📁 server/               # Express.js Backend
│       ├── index.js             #   30+ API endpoints
│       ├── stock-service.js     #   East Money data fetcher
│       ├── websocket-service.js #   Real-time push service
│       └── 📁 data/db.json      #   LowDB JSON storage
│
├── 📁 admin/                    # Admin Panel (Vue 3)
├── 📁 baostock/                 # BaoStock Python API (FastAPI)
├── 📁 akshare/                  # AKShare Python API (FastAPI)
├── 📁 ml_services/              # ML Inference (backtest, prediction)
├── 📁 docs/                     # 20+ documentation files
└── 📁 configs/                  # Training configs

📊 Feature Comparison

Feature	Web App	Desktop App	Notes
📈 Real-time Market Data	✅	✅	3-source fallback on desktop
🤖 AI Chat (20+ models)	—	✅	Desktop exclusive
🎯 Model Selector UI	—	✅	Desktop exclusive
📎 File Upload in Chat	—	✅	Desktop exclusive
🎙️ Voice Input	—	✅	Desktop exclusive
💼 Portfolio Management	✅	✅
⭐ Watchlist	✅	✅
📊 Index Valuation	✅	✅
📦 ETF Market	✅	✅
🧪 Strategy Backtesting	✅	✅	5 strategies: MA/MACD/KDJ/RSI/BB
🔍 Smart Stock Search	✅	✅
🎨 Theme Customization	—	✅	6 colors + dark/light/auto
🌍 Multi-language	—	✅	CN/TW/EN/JP
📱 System Tray	—	✅	Desktop exclusive
🎛️ Admin Panel	✅	—	Web exclusive

📝 Changelog

v1.0.0 (2026-02-24) — In Fortune AI Release

Note

Project renamed from JiuCai (久财) to In Fortune AI. Complete codebase rebranding across 76+ files.

🤖 Multi-Model AI

20+ models across 5 categories (Dialogue, Search, Tool-calling, GPT/Grok, Image Gen)
In-chat model selector with categorized dropdown
Backend /api/chat now accepts dynamic model parameter

📈 Market Data Integration

BaoStock direct connection for A-share index K-line data
East Money free API for real-time index quotes
3-tier fallback architecture (Backend → BaoStock → East Money)

💬 Chat Enhancements

File upload support (images, PDF, Word, Excel, CSV, Markdown)
Voice input via Web Speech API
Chat history with timestamps, persisted to localStorage

🖥️ Desktop Fixes

Fixed Windows keyboard input/IME in Electron (menu + focus fixes)
Updated CSP for multi-origin data access
Increased API timeout to 60s, max_tokens to 8000

🤝 Contributing

Fork this repository
Create a feature branch: git checkout -b feature/amazing-feature
Commit changes: git commit -m 'Add amazing feature'
Push: git push origin feature/amazing-feature
Open a Pull Request

Tip

Check docs/INDEX.md for the full documentation index with 20+ guides.

🙏 Credits

@QoneFeng
Original InFortune.ai project

@sjkncs
Desktop app & multi-model integration

Built with: Electron · Vue.js · FastAPI · BaoStock · AKShare · TailwindCSS · ECharts

📄 License

This project is licensed under the MIT License.

Made with ❤️ by Yangting SONG

_{Mathematics is the language of nature}

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
admin		admin
akshare		akshare
baostock		baostock
configs		configs
data_layer		data_layer
desktop-app		desktop-app
docs		docs
ml_services		ml_services
scripts		scripts
tests		tests
website		website
.env.example		.env.example
.gitignore		.gitignore
EVALUATION_REPORT.md		EVALUATION_REPORT.md
INSTALL_ALL.bat		INSTALL_ALL.bat
README.md		README.md
README_FIRST.md		README_FIRST.md
RELEASE_DESCRIPTION.md		RELEASE_DESCRIPTION.md
START_ALL.bat		START_ALL.bat
build-windows.bat		build-windows.bat
docker-compose.research.yml		docker-compose.research.yml
requirements-research.txt		requirements-research.txt
setup-github.ps1		setup-github.ps1

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Repository files navigation

🚀 What's New in v1.0.0

🤖 Multi-Model AI Engine

📈 Real-Time Market Data

💬 Enhanced Chat Interface

🖥️ Desktop App Improvements

🤖 Multi-Model AI Engine

📈 Real-Time Market Data

🖥️ Desktop Application

10 View Modules

Chat Interface Features

🏗️ Architecture

Service Ports

⚡ Quick Start

Prerequisites

1. Clone & Install

2. Configure

3. Run

4. Build for Distribution

📡 API Reference

📁 Project Structure

📊 Feature Comparison

📝 Changelog

v1.0.0 (2026-02-24) — In Fortune AI Release

🤝 Contributing

🙏 Credits

📄 License

About

Uh oh!

Releases 1

Packages

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Folders and files

Latest commit

History

Repository files navigation

🚀 What's New in v1.0.0

🤖 Multi-Model AI Engine

📈 Real-Time Market Data

💬 Enhanced Chat Interface

🖥️ Desktop App Improvements

🤖 Multi-Model AI Engine

📈 Real-Time Market Data

🖥️ Desktop Application

10 View Modules

Chat Interface Features

🏗️ Architecture

Service Ports

⚡ Quick Start

Prerequisites

1. Clone & Install

2. Configure

3. Run

4. Build for Distribution

📡 API Reference

📁 Project Structure

📊 Feature Comparison

📝 Changelog

v1.0.0 (2026-02-24) — In Fortune AI Release

🤝 Contributing

🙏 Credits

📄 License

About

Topics

Resources

Uh oh!

Stars

Watchers

Forks

Releases 1

Packages 0

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Packages