aigverse - A Python Library for Logic Networks, Synthesis, and Optimization¶
aigverse is an open-source C++17 and Python infrastructure library for working with logic networks, synthesis, and
optimization in Python-first workflows.
It builds directly upon the EPFL Logic Synthesis Libraries, particularly mockturtle, kitty, and lorina, and exposes these mature C++ capabilities through an idiomatic Python interface. This allows users to run synthesis workflows from Python without reimplementing core algorithms there. The result is reusable support for And-Inverter Graph (AIG) construction and manipulation, optimization and equivalence checking, dataset generation, and export into graph and numeric representations for downstream data science and ML pipelines.
Key features include:
Efficient logic representation using And-Inverter Graphs (AIGs)
Support for various file formats (AIGER, Verilog, Bench, PLA)
High-performance C++ backend with a Pythonic interface
Optional adapters for graph and numeric interoperability in machine learning and data science workflows
Comprehensive tools for logic synthesis and optimization
aigverse is designed as reusable bridge infrastructure rather than as a full end-to-end EDA toolchain.
This documentation provides a comprehensive guide to the aigverse library, including {doc}
installation instructions <installation>, a quickstart guide, and detailed {doc}
API documentation <api/aigverse/index>.
The source code of aigverse is publicly available on GitHub
at marcelwa/aigverse, while pre-built binaries are available
via PyPI for all major operating systems and all modern Python versions.
See also
For a deeper dive into the vision and technical details behind aigverse, see the presentation “aigverse: Toward machine learning-driven logic synthesis” from the Free Silicon Conference (FSiC) 2025. The slides are available on the FSiC wiki and cover the motivation, architecture, and future directions of the aigverse project.
User Guide
Developers