qusong0627/QuantMind
QuantMind（量化大脑）开源版是一款面向个人开发者与投研团队的 AI 原生多市场量化交易平台。深度集成微软 Qlib、RD-Agent 因子演化与 TradingAgents 多智能体投研，提供从 300+ 维因子挖掘、13 种机器学习与深度学习模型工场、Optuna 自动调参、Qlib 高性能回测、截面批量推理、7x24 实时舆情情绪分析，到通达信深度联动（板块推送/预警雷达/闪电下单）与实盘模拟交易的完整闭环。全面支持 A股、港股、美股、期货及区块链。系统采用 Docker Compose 一键私有化部署，数据与模型完全本地化，保障策略隐私，功能零门槛无限制。
What it solves
QuantMind is an all-in-one AI-native quantitative trading platform designed for individual researchers and professional institutions. It eliminates the need for manual data cleaning and complex code assembly by automating the entire quantitative workflow—from data acquisition and factor mining to model training, backtesting, and live trading execution across A-shares, Hong Kong, US stocks, futures, and blockchain markets.
How it works
The platform integrates several high-performance frameworks to create a closed-loop system:
- Data & Analysis: Uses a data hub with 300+ pre-computed features, utilizing Parquet and DuckDB for high-speed columnar storage and retrieval.
- Factor Evolution: Employs Microsoft's RD-Agent (AutoAlpha 2.0) to allow users to describe quantitative hypotheses in natural language, which the AI then converts into formulas and evolves through backtesting.
- Model Factory: Provides 13 different machine learning and deep learning models (including LightGBM, Transformer, and TabNet) with automated hyperparameter tuning via Optuna.
- Execution: Connects to brokerage systems like MiniQMT (via a WebSocket Bridge) and integrates with TongdaXing for real-time alerts and one-click ordering.
- Backtesting: Uses the Microsoft Qlib engine for high-performance event-driven backtesting and risk attribution.
Who it’s for
- Individual quantitative researchers looking to automate their alpha discovery.
- Investment research teams needing an integrated pipeline for model training and deployment.
- Professional traders who want to bridge AI-driven signals with live brokerage execution.
Highlights
- AI-Driven Factor Mining: Natural language to factor formula synthesis via LLM-driven evolution.
- Extensive Model Library: Support for 13 ML/DL algorithms with seamless GPU scheduling (local or AutoDL).
- Live Trading Integration: Native support for MiniQMT and TongdaXing for real-world execution.
- Quantitative Data Hub: Built-in 300+ dimensional features and 7x24 RSS sentiment monitoring for news-based alpha.
Related
- Project
- Project
- Project
- Project
- Project