akfamily/akquant

Health v2 · maintenance only

Adoption not part of the score

2095 stars · 274 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

AKQuant is a high-performance quantitative research and backtesting framework with a Rust core engine and Python interface for strategy development. It provides event-driven backtesting, machine learning integration (walk-forward validation with PyTorch/Scikit-learn), a factor expression engine, and built-in risk management.

Use cases

• backtest trading strategies on historical stock data
• run walk-forward validation for ML-based trading strategies
• compute Alpha101-style factor expressions with Polars
• optimize strategy parameters with parallel grid search
• backtest multi-asset portfolios with risk controls
• fetch A-share market data via akshare and run backtests
• generate benchmark comparison reports for strategies

When to choose

• you need fast event-driven backtesting with a Rust-optimized engine
• you want to combine machine learning models with trading strategy backtests
• you work with Chinese A-share market data via akshare
• you need built-in technical indicators, factor computation, and parameter optimization in one framework

When to avoid

• you need live trading execution against real brokers rather than backtesting
• you require a large community ecosystem like Zipline or Backtrader with extensive third-party examples
• you need asset classes or markets beyond what the framework's data integrations support

Facets

framework · maturity active

simulation machine-learning data-science benchmarking sdk fintech quantum-computing machine-learning data-science python rust cross-platform windows quantitative-finance backtesting trading-strategies event-driven walk-forward-validation factor-analysis akshare rust-core technical-indicators parameter-optimization macos linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="akfamily/akquant")