Open Source
Portfolio Optimization & Factor Risk Models
Production-ready Python infrastructure for portfolio construction, custom risk models, and performance attribution, backed by enterprise support
Built for Quantitative Finance
From quant hedge funds and asset managers to AI innovators, skfolio powers diverse applications across the industry
Portfolio Managers
Custom factor risk models, alpha decomposition, factor-neutral optimization and attribution in one pipeline
Index & QIS Teams
Systematic strategy design with pre-selection, tail-risk optimization and scenario generation in a unified pipeline
AI-Driven Investment
Agentic AI with LLM context and MCP for orchestrated workflows, accelerating research and deployment
Fintech & DeFi
Automation and advanced models for faster innovation, strategy execution and adaptability in dynamic markets
Explore 100+ Models
From custom factor risk models and optimization to stress testing and synthetic data generation, discover the breadth of skfolio’s capabilities
>>> >>> >>> >>> >>> >>> From Open Source to Enterprise
An open source foundation, extended by enterprise support and industry expertise from Skfolio Labs
Open Source Library
Permissive license, no vendor lock-in and fully auditable for infrastructure that institutions can trust and extend
Production-Ready
Institutional reliability with CI/CD pipelines, 13,000+ unit tests and peer review ensuring production stability
Modular & Interoperable
Built on the scikit-learn API, data-agnostic and compatible with any commercial optimizer, factor model and data vendor
Enterprise Support
SLAs, bespoke development and roadmap access, providing stability and alignment with institutional requirements
Build on a Trusted API
See how skfolio takes you from initial experiments to complex pipelines in just a few lines of code, leveraging the scikit-learn paradigm
Expert Voices in Quant Finance
Feedback from leading researchers and practitioners using skfolio
skfolio is clean and well-managed. By adhering closely to scikit-learn conventions, one gets clean code for things like statistical stacking of portfolio methods. It now includes a way to compute Schur portfolios that you won’t find anywhere else.
skfolio is a real game-changer for portfolio optimization in Python. It combines production-grade tools with a remarkably intuitive, scikit-learn-compatible interface, making complex workflows accessible and efficient. What sets it apart is the breadth and depth of its optimization designs, the flexibility in constraints and factor-neutralization settings, as well as the seamless choice between open-source and professional solvers. Add to that robust backtesting capabilities, and you get an open-source package that in many ways surpasses most commercial solutions.
Python's skfolio library provides an elegant, unified, scikit-learn-compatible framework for portfolio optimization that seamlessly integrates a wide variety of portfolio designs, backtesting capabilities (from walk-forward to multiple randomized backtests), and comprehensive visualization. The entire workflow couldn't be more straightforward for quantitative researchers and practitioners.
Modular & Extensible
Build, adapt and integrate across the entire portfolio pipeline, assembling components into flexible workflows
Encode distributional assumptions using empirical estimates, Black-Litterman, entropy or opinion pooling, and factor risk models. Build custom fundamental cross-sectional factor models from point-in-time characteristics with style, industry, country and currency factors, alpha decomposition and stressed synthetic scenarios.
Open Source to Partnership
Whether exploring, scaling, or innovating, Skfolio Labs offers a pathway for every stage of adoption
Open Source
The core library, freely available
- BSD-3 license, full source code access
- Documentation, examples and tutorials
- Community discussions and contributions
Enterprise
Support and expertise for production
- SLAs and escalation paths
- Dedicated technical support channels
- Roadmap access and priority input
- Integration guidance for data vendors, optimizers and factor models
Tailored
Customized solutions for advanced teams
- Bespoke feature development and integration
- Advisory on risk methodologies and optimization design
- Dedicated account management with custom SLAs
- Strategic partnerships for joint initiatives
From Exploration to Enterprise
Start with our comprehensive documentation or connect with us to discuss enterprise solutions