QuantDinger

🇺🇸 English | 🇨🇳 简体中文 | 繁體中文 | 🇯🇵 日本語 | 🇰🇷 한국어

Next-Gen AI Quantitative Trading Platform

🤖 AI-Native · 🐍 Visual Python · 🌍 Multi-Market · 🔒 Privacy-First

Build, Backtest, and Trade with an AI Co-Pilot. Better than PineScript, Smarter than SaaS.

Official Community · Live Demo · 📺 Video Demo · 🌟 Join Us

---

📖 Introduction

What is QuantDinger?

QuantDinger is a local-first, privacy-first, self-hosted quantitative trading infrastructure. It runs on your own machine/server, providing multi-user accounts backed by PostgreSQL while keeping full control of your strategies, trading data, and API keys.

Why Local-First?

Unlike SaaS platforms that lock your data and strategies in the cloud, QuantDinger runs locally. Your strategies, trading logs, API keys, and analysis results stay on your machine. No vendor lock-in, no subscription fees, no data exfiltration.

Who is this for?

QuantDinger is built for traders, researchers, and engineers who:

• Value data sovereignty and privacy
• Want transparent, auditable trading infrastructure
• Prefer engineering over marketing
• Need a complete workflow: data, analysis, backtesting, and execution

Core Features

QuantDinger includes a built-in LLM-based multi-agent research system that gathers financial intelligence from the web, combines it with local market data, and generates analysis reports. This integrates with strategy development, backtesting, and live trading workflows.

Core Value

• 🔓 Apache 2.0 Open Source (Code): Permissive and commercial-friendly. You can fork and modify the codebase under Apache 2.0, while preserving required notices.
• 🐍 Python-Native & Visual: Write indicators in standard Python (easier than PineScript) with AI assistance. Visualize signals directly on charts—a "Local TradingView" experience.
• 🤖 AI-Loop Optimization: It doesn't just run strategies; AI analyzes backtest results to suggest parameter tuning (Stop-Loss/TP/MACD settings), forming a closed optimization loop.
• 🌍 Universal Market Access: One unified system for Crypto (Live), US/CN Stocks, Forex, and Futures (Data/Notify).
• ⚡ Docker & Clean Arch: 4-line command deployment. Modern Tech Stack (Vue + Python) with a clean, separation-of-concerns architecture.

---

📺 Video Demo

Click the video above to watch the QuantDinger project introduction

---

📚 Documentation

Guides

• Python Strategy Development Guide
• Interactive Brokers (IBKR) Trading Guide 🆕
• MetaTrader 5 (MT5) Trading Guide 🆕

Notification Configuration

• Telegram Notification Setup
• Email (SMTP) Notification Setup
• SMS (Twilio) Notification Setup

📸 Visual Tour

🗺️ System Architecture Overview

A comprehensive view of QuantDinger's AI-powered research, backtesting, and automated trading capabilities.

📊 Professional Quant Dashboard

Real-time monitoring of market dynamics, assets, and strategy status.

🤖 AI Deep Research Multi-agent collaboration for market sentiment & technical analysis.	💬 Smart Trading Assistant Natural language interface for instant market insights.
📈 Interactive Indicator Analysis Rich library of technical indicators with drag-and-drop analysis.	🐍 Python Strategy Gen Built-in editor with AI-assisted strategy coding.
📊 Portfolio Monitor Track positions, set alerts, and receive AI-powered analysis via Email/Telegram.

---

✨ Key Features

1. Visual Python Strategy Workbench

Better than PineScript, Smarter than SaaS.

• Python Native: Write indicators and strategies in Python. Leverage the entire Python ecosystem (Pandas, Numpy, TA-Lib) instead of proprietary languages like PineScript.
• "Mini-TradingView" Experience: Run your Python indicators directly on the built-in K-line charts. Visually debug buy/sell signals on historical data.
• AI-Assisted Coding: Let the built-in AI write the complex logic for you. From idea to code in seconds.

2. Complete Trading Lifecycle

From Indicator to Execution, Seamlessly.

1. Indicator: Define your market entry/exit signals.
2. Strategy Config: Attach risk management rules (Position sizing, Stop-Loss, Take-Profit).
3. Backtest & AI Optimization: Run backtests, view rich performance metrics, and let AI analyze the result to suggest improvements (e.g., "Adjust MACD threshold to X").
4. Execution Mode:
   • Live Trading:
     • Cryptocurrency: Direct API execution for 10+ exchanges (Binance, OKX, Bitget, Bybit, etc.)
     • US/HK Stocks: Via Interactive Brokers (IBKR) 🆕
     • Forex: Via MetaTrader 5 (MT5) 🆕
   • Signal Notification: For markets without live trading support (A-shares/Futures), send signals via Telegram, Discord, Email, SMS, or Webhook.

3. AI-Powered Analysis

Fast, Accurate, Professional Reports.

QuantDinger features a streamlined AI analysis system:

• Fast Analysis Mode: Single LLM call architecture for quick, accurate analysis (replaces complex multi-agent system)
• Global Market Integration: Real-time market data, heatmaps, and economic calendar integrated into the analysis page
• ATR-Based Trading Levels: Stop-loss and take-profit recommendations based on technical analysis (ATR, Support/Resistance)
• Analysis Memory: Store analysis results for history review and continuous learning
• Strategic Integration: AI analysis can serve as a "Market Filter" for your strategies

4. Universal Data Engine

QuantDinger provides a unified data interface across multiple markets:

• Cryptocurrency: Direct API connections for trading (10+ exchanges) and CCXT integration for market data (100+ sources)
• Stocks: Yahoo Finance, Finnhub, Tiingo (US stocks), and AkShare (CN/HK stocks)
• Futures/Forex: OANDA and major futures data sources
• Proxy Support: Built-in proxy configuration for restricted network environments

5. Memory-Augmented Agents (Local RAG + Reflection Loop)

QuantDinger’s agents don’t start from scratch every time. The backend includes a local memory store and an optional reflection/verification loop:

• What it is: RAG-style experience retrieval injected into agent prompts (NOT model fine-tuning).
• Where it lives: PostgreSQL database (shared with main data) or local files under backend_api_python/data/memory/ (privacy-first).

flowchart TB
    %% ===== 🌐 Entry Layer =====
    subgraph Entry["🌐 API Entry"]
        A["📡 POST /api/analysis/multi"]
        A2["🔄 POST /api/analysis/reflect"]
    end

    %% ===== ⚙️ Service Layer =====
    subgraph Service["⚙️ Service Orchestration"]
        B[AnalysisService]
        C[AgentCoordinator]
        D["📊 Build Context<br/>price · kline · news · indicators"]
    end

    %% ===== 🤖 Multi-Agent Workflow =====
    subgraph Agents["🤖 Multi-Agent Workflow"]

        subgraph P1["📈 Phase 1 · Analysis (Parallel)"]
            E1["🔍 MarketAnalyst<br/><i>Technical</i>"]
            E2["📑 FundamentalAnalyst<br/><i>Fundamentals</i>"]
            E3["📰 NewsAnalyst<br/><i>News & Events</i>"]
            E4["💭 SentimentAnalyst<br/><i>Market Mood</i>"]
            E5["⚠️ RiskAnalyst<br/><i>Risk Assessment</i>"]
        end

        subgraph P2["🎯 Phase 2 · Debate (Parallel)"]
            F1["🐂 BullResearcher<br/><i>Bullish Case</i>"]
            F2["🐻 BearResearcher<br/><i>Bearish Case</i>"]
        end

        subgraph P3["💹 Phase 3 · Decision"]
            G["🎰 TraderAgent<br/><i>Final Verdict → BUY / SELL / HOLD</i>"]
        end

    end

    %% ===== 🧠 Memory Layer =====
    subgraph Memory["🧠 PostgreSQL Memory Store"]
        M1[("market_analyst")]
        M2[("fundamental")]
        M3[("news_analyst")]
        M4[("sentiment")]
        M5[("risk_analyst")]
        M6[("bull_researcher")]
        M7[("bear_researcher")]
        M8[("trader_agent")]
    end

    %% ===== 🔄 Reflection Loop =====
    subgraph Reflect["🔄 Reflection Loop (Optional)"]
        R[ReflectionService]
        RR[("reflection_records.db")]
        W["⏰ ReflectionWorker"]
    end

    %% ===== Main Flow =====
    A --> B --> C --> D
    D --> P1 --> P2 --> P3

    %% ===== Memory Read/Write =====
    E1 <-.-> M1
    E2 <-.-> M2
    E3 <-.-> M3
    E4 <-.-> M4
    E5 <-.-> M5
    F1 <-.-> M6
    F2 <-.-> M7
    G <-.-> M8

    %% ===== Reflection Flow =====
    C --> R --> RR
    W --> RR
    W -.->|"verify + learn"| M8
    A2 -.->|"manual review"| M8

Retrieval ranking (simplified):

[ score = w_{sim}\cdot sim + w_{recency}\cdot recency + w_{returns}\cdot returns_score ]

Config lives in .env (see backend_api_python/env.example): ENABLE_AGENT_MEMORY, AGENT_MEMORY_TOP_K, AGENT_MEMORY_ENABLE_VECTOR, AGENT_MEMORY_HALF_LIFE_DAYS, and ENABLE_REFLECTION_WORKER.

6. Strategy Runtime

• Thread-Based Executor: Independent thread pool for strategy execution
• Auto-Restore: Resumes running strategies after system restarts
• Order Queue: Background worker for order execution

7. Multi-LLM Provider Support

QuantDinger supports multiple AI providers with auto-detection:

Provider	Features
OpenRouter	Multi-model gateway (default), 100+ models
OpenAI	GPT-4o, GPT-4o-mini
Google Gemini	Gemini 1.5 Flash/Pro
DeepSeek	DeepSeek Chat (cost-effective)
xAI Grok	Grok Beta

Simply configure your preferred provider's API key in .env. The system auto-detects available providers.

8. Indicator Community

Share, Discover, and Trade Indicators.

• Publish & Share: Share your Python indicators with the community
• Purchase System: Buy premium indicators from other users
• Rating & Reviews: Rate and review purchased indicators
• Admin Review: Moderation system for quality control

9. User Management & Security

• Multi-User Support: PostgreSQL-backed user accounts with role-based permissions
• OAuth Login: Google and GitHub OAuth integration
• Email Verification: Registration and password reset via email codes
• Security Features: Cloudflare Turnstile captcha, IP/account rate limiting
• Demo Mode: Read-only mode for public demonstrations

10. Tech Stack

• Backend: Python (Flask) + PostgreSQL + Redis (optional)
• Frontend: Vue 2 + Ant Design Vue + KlineCharts/ECharts
• Deployment: Docker Compose (with PostgreSQL)
• Current Version: V2.1.1 (Changelog)

---

🔌 Supported Exchanges & Brokers

QuantDinger supports multiple execution methods for different market types:

Cryptocurrency Exchanges (Direct API)

Exchange	Markets
Binance	Spot, Futures, Margin
OKX	Spot, Perpetual, Options
Bitget	Spot, Futures, Copy Trading
Bybit	Spot, Linear Futures
Coinbase Exchange	Spot
Kraken	Spot, Futures
KuCoin	Spot, Futures
Gate.io	Spot, Futures
Bitfinex	Spot, Derivatives

Traditional Brokers

Broker	Markets	Platform
Interactive Brokers (IBKR)	US Stocks, HK Stocks	TWS / IB Gateway 🆕
MetaTrader 5 (MT5)	Forex	MT5 Terminal 🆕

Market Data (via CCXT)

Bybit, Gate.io, Kraken, KuCoin, HTX, and 100+ other exchanges for market data.

---

Multi-Language Support

QuantDinger is built for a global audience with comprehensive internationalization:

All UI elements, error messages, and documentation are fully translated. Language is auto-detected based on browser settings or can be manually switched in the app.

---

Supported Markets

Market Type	Data Sources	Trading
Cryptocurrency	Binance, OKX, Bitget, + 100 exchanges	✅ Full support
US Stocks	Yahoo Finance, Finnhub, Tiingo	✅ Via IBKR 🆕
HK Stocks	AkShare, East Money	✅ Via IBKR 🆕
CN Stocks (A-shares)	AkShare, East Money	⚡ Data only
Forex	Finnhub, OANDA	✅ Via MT5 🆕
Futures	Exchange APIs, AkShare	⚡ Data only

---

Architecture (Current Repo)

┌─────────────────────────────┐
│      quantdinger_vue         │
│   (Vue 2 + Ant Design Vue)   │
└──────────────┬──────────────┘
               │  HTTP (/api/*)
               ▼
┌─────────────────────────────┐
│     backend_api_python       │
│   (Flask + strategy runtime) │
└──────────────┬──────────────┘
               │
               ├─ PostgreSQL (multi-user support)
               ├─ Redis (optional cache)
               └─ Data providers / LLMs / Exchanges

---

Repository Layout

.
├─ backend_api_python/         # Flask API + AI + backtest + strategy runtime
│  ├─ app/
│  ├─ env.example              # Copy to .env for local config
│  ├─ requirements.txt
│  └─ run.py                   # Entrypoint
└─ quantdinger_vue/            # Vue 2 UI (dev server proxies /api -> backend)

---

Quick Start

Option 1: Docker Deployment (Recommended)

The fastest way to get QuantDinger running with PostgreSQL database and multi-user support.

1. Configure Environment

Create a .env file in project root:

# Database Configuration
POSTGRES_USER=quantdinger
POSTGRES_PASSWORD=your_secure_password
POSTGRES_DB=quantdinger

# Admin Account (created on first startup)
ADMIN_USER=quantdinger
ADMIN_PASSWORD=123456

# Optional: AI Features
OPENROUTER_API_KEY=your_api_key

2. Start Services

Linux / macOS

git clone https://github.com/brokermr810/QuantDinger.git && \
cd QuantDinger && \
cp backend_api_python/env.example backend_api_python/.env && \
docker-compose up -d --build

Windows (PowerShell)

git clone https://github.com/brokermr810/QuantDinger.git
cd QuantDinger
Copy-Item backend_api_python\env.example -Destination backend_api_python\.env
docker-compose up -d --build

This will automatically:

• Start PostgreSQL database (port 5432)
• Initialize database schema
• Start backend API (port 5000)
• Start frontend (port 8888)
• Create admin user from ADMIN_USER/ADMIN_PASSWORD in .env

3. Access the Application

• Frontend UI: http://localhost:8888
• Backend API: http://localhost:5000
• Default Account: Uses ADMIN_USER / ADMIN_PASSWORD from .env (default: quantdinger / 123456, please change for production)

Note

: For production, edit backend_api_python/.env to set strong passwords, add OPENROUTER_API_KEY for AI features, then restart with docker-compose restart backend.

Docker Commands Reference

# View running status
docker-compose ps

# View logs
docker-compose logs -f

# View backend logs only
docker-compose logs -f backend

# View frontend logs only
docker-compose logs -f frontend

# Stop services
docker-compose down

# Stop and remove volumes (WARNING: deletes database!)
docker-compose down -v

# Restart services
docker-compose restart

# Rebuild and restart
docker-compose up -d --build

# Enter backend container
docker exec -it quantdinger-backend /bin/bash

# Enter frontend container
docker exec -it quantdinger-frontend /bin/sh

Docker Architecture

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│   Frontend      │     │    Backend      │     │   PostgreSQL    │
│   (Nginx)       │────▶│   (Python)      │────▶│   Database      │
│   Port: 8888    │     │   Port: 5000    │     │   Port: 5432    │
└─────────────────┘     └─────────────────┘     └─────────────────┘
        │                       │                       │
        └───────────────────────┴───────────────────────┘
                         Docker Network

• Frontend: Vue.js app served by Nginx, proxies API requests to backend
• Backend: Python Flask API service with multi-user authentication
• PostgreSQL: Database for user data, strategies, and trading records

Data Persistence

The following data is persisted across container restarts:

volumes:
  postgres_data:                                            # PostgreSQL database
  - ./backend_api_python/logs:/app/logs                     # Logs
  - ./backend_api_python/data:/app/data                     # Data directory
  - ./backend_api_python/.env:/app/.env                     # Configuration

Customization

Change ports - Edit docker-compose.yml:

services:
  frontend:
  ports:
    - "8080:80"  # Change to port 8080
  
  backend:
  ports:
    - "5001:5000"  # Change to port 5001

Configure HTTPS - Use a reverse proxy (like Caddy/Nginx):

# Using Caddy (automatic HTTPS)
caddy reverse-proxy --from yourdomain.com --to localhost:80

Production Recommendations

Security:

# Generate strong SECRET_KEY
openssl rand -hex 32

# Set secure admin password
ADMIN_PASSWORD=your-very-secure-password

Resource limits - Add to docker-compose.yml:

services:
  backend:
  deploy:
    resources:
    limits:
      cpus: '2'
      memory: 2G
    reservations:
      cpus: '0.5'
      memory: 512M

Log management:

services:
  backend:
  logging:
    driver: "json-file"
    options:
    max-size: "100m"
    max-file: "3"

Docker Troubleshooting

Frontend can't connect to backend:

docker-compose logs backend
curl http://localhost:5000/api/health

Database connection issues:

# Check PostgreSQL container status
docker-compose logs postgres

# Verify PostgreSQL is ready
docker exec quantdinger-db pg_isready -U quantdinger

# Connect to database manually
docker exec -it quantdinger-db psql -U quantdinger -d quantdinger

Build failures:

# Clear Docker cache and rebuild
docker-compose build --no-cache

Out of memory:

# Check memory usage
docker stats

# Add swap space (Linux)
sudo fallocate -l 2G /swapfile
sudo chmod 600 /swapfile
sudo mkswap /swapfile
sudo swapon /swapfile

Updating

# Pull latest code
git pull

# Rebuild and restart
docker-compose up -d --build

Backup

# Backup PostgreSQL database
docker exec quantdinger-db pg_dump -U quantdinger quantdinger > backup/quantdinger_$(date +%Y%m%d).sql

# Backup configuration
cp backend_api_python/.env backup/.env_$(date +%Y%m%d)

# Restore database (if needed)
cat backup/quantdinger_YYYYMMDD.sql | docker exec -i quantdinger-db psql -U quantdinger quantdinger

---

Option 2: Local Development

Prerequisites

• Python 3.10+ recommended
• Node.js 16+ recommended
• PostgreSQL 14+ installed and running

1. Setup PostgreSQL

# Create database and user
sudo -u postgres psql
CREATE DATABASE quantdinger;
CREATE USER quantdinger WITH ENCRYPTED PASSWORD 'your_password';
GRANT ALL PRIVILEGES ON DATABASE quantdinger TO quantdinger;
\q

# Initialize schema
psql -U quantdinger -d quantdinger -f backend_api_python/migrations/init.sql

2. Start the backend (Flask API)

cd backend_api_python
pip install -r requirements.txt
cp env.example .env   # Windows: copy env.example .env

Edit .env and set:

DATABASE_URL=postgresql://quantdinger:your_password@localhost:5432/quantdinger
SECRET_KEY=your-secret-key
ADMIN_USER=quantdinger
ADMIN_PASSWORD=123456

Then start:

python run.py

Backend will be available at http://localhost:5000.

2. Start the frontend (Vue UI)

cd quantdinger_vue
npm install
npm run serve

Frontend dev server runs at http://localhost:8000 and proxies /api/* to http://localhost:5000 (see quantdinger_vue/vue.config.js).

---

Configuration (.env)

Use backend_api_python/env.example as a template. Common settings include:

• Auth: SECRET_KEY, ADMIN_USER, ADMIN_PASSWORD
• Server: PYTHON_API_HOST, PYTHON_API_PORT, PYTHON_API_DEBUG
• Database: DATABASE_URL (PostgreSQL connection string)
• AI / LLM: LLM_PROVIDER (openrouter/openai/google/deepseek/grok), provider-specific API keys
• OAuth: GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, GITHUB_CLIENT_ID, GITHUB_CLIENT_SECRET
• Security: TURNSTILE_SITE_KEY, TURNSTILE_SECRET_KEY, ENABLE_REGISTRATION
• Web search: SEARCH_PROVIDER, SEARCH_GOOGLE_*, SEARCH_BING_API_KEY
• Order Execution: ORDER_MODE (maker/market), MAKER_WAIT_SEC, MAKER_OFFSET_BPS
• Proxy (optional): PROXY_PORT or PROXY_URL
• Workers: ENABLE_PENDING_ORDER_WORKER, ENABLE_PORTFOLIO_MONITOR

---

API

The backend provides REST endpoints for login, market data, indicators, backtesting, strategies, and AI analysis.

• Health: GET /health (also supports GET /api/health for deployment probes)
• Auth (frontend-compatible): POST /api/user/login, POST /api/user/logout, GET /api/user/info

For the full route list, see backend_api_python/app/routes/.

---

License

Licensed under the Apache License 2.0. See LICENSE.

---

🤝 Community & Support

• Contributing: Contributing Guide · Contributors
• Telegram: QuantDinger Group
• Discord: Join Server
• 📺 Video Demo: Project Introduction
• YouTube: @quantdinger
• Email: brokermr810@gmail.com
• GitHub Issues: Report bugs / Request features

---

💼 Commercial License & Sponsorship

QuantDinger is licensed under Apache License 2.0 (code). However, Apache 2.0 does NOT grant trademark rights. Our branding assets (name/logo) are protected as trademarks and are governed separately from the code license:

• Copyright/Attribution: You must keep required copyright and license notices (including any NOTICE/attribution in the repo and in the UI where applicable).
• Trademarks (Name/Logo/Branding): Without permission, you may not modify QuantDinger branding (name/logo/UI brand), or use it to imply endorsement or misrepresent origin. If you redistribute a modified version, you should remove QuantDinger branding and rebrand unless you have a commercial license.

If you need to keep/modify QuantDinger branding in a redistribution (including UI branding and logo usage), please contact us for a commercial license.

See: TRADEMARKS.md

What you get with a Commercial License

• Commercial authorization to modify branding/copyright display as agreed
• Operations support: deployment, upgrades, incident support, and maintenance guidance
• Consulting services: architecture review, performance tuning, strategy workflow consulting
• Sponsorship options: become a project sponsor and we can display your logo/ad (README/website/in-app placement as agreed)

Contact

• Telegram: https://t.me/worldinbroker
• Email: brokermr810@gmail.com

---

💝 Direct Support (Donations)

Your contributions help us maintain and improve QuantDinger.

Crypto Donations (ERC-20 / BEP-20 / Polygon / Arbitrum)

0x96fa4962181bea077f8c7240efe46afbe73641a7

---

🎓 Supporting Partners

We are proud to be supported by academic institutions and organizations advancing quantitative finance education and research.

Quantitative Finance Society (QFS) Indiana University Bloomington Fostering the next generation of quantitative finance professionals

💡 Interested in becoming a supporting partner? We welcome collaborations with universities, research institutions, and organizations. Contact us at brokermr810@gmail.com or via Telegram.

---

Acknowledgements

QuantDinger stands on the shoulders of great open-source projects:

Project	Description	Link
Flask	Lightweight WSGI web framework	flask.palletsprojects.com
flask-cors	Cross-Origin Resource Sharing extension	GitHub
Pandas	Data analysis and manipulation library	pandas.pydata.org
CCXT	Cryptocurrency exchange trading library	github.com/ccxt/ccxt
yfinance	Yahoo Finance market data downloader	github.com/ranaroussi/yfinance
akshare	China financial data interface	github.com/akfamily/akshare
requests	HTTP library for Python	requests.readthedocs.io
Vue.js	Progressive JavaScript framework	vuejs.org
Ant Design Vue	Enterprise-class UI components	antdv.com
KlineCharts	Lightweight financial charting library	github.com/klinecharts/KLineChart
Lightweight Charts	TradingView charting library	github.com/nicepkg/lightweight-charts
ECharts	Apache data visualization library	echarts.apache.org

Thanks to all maintainers and contributors across these ecosystems! ❤️