Installation | Documentation | Examples/Tutorials | FAQs | Citing GWKokab
GWKokab is a high-performance, flexible, and easy-to-use toolkit for gravitational-wave population inference. Built on top of JAX, it enables efficient Bayesian inference for a wide range of parametric population models while remaining fully compatible with modern GPU/TPU-accelerated workflows.
The framework is designed to support scalable hierarchical inference and rapid experimentation with astrophysical population models, including mass, spin, redshift, and eccentricity distributions of compact binary mergers.
Important
The latest tested features, updates, and bug fixes are currently available on the dev branch.
Until the ongoing documentation updates are finalized, we recommend users install and work from the dev branch instead of main.
Clone directly using:
git clone -b dev https://github.com/kokabsc/gwkokab.gitor switch an existing clone:
git checkout dev
git pull origin devThe main branch will be updated after the current development cycle and documentation for the dev branch are completed.
[!NOTE]
For new users, we recommend starting with the NumPyro sampler before using FlowMC.
NumPyro is generally easier to configure, debug, and tune, making it a more accessible starting point for developing and validating population inference workflows.
We welcome contributions from the community.
If you would like to contribute to GWKokab, please see the
contributing guidelines.
If you use GWKokab in your research, please cite the following works:
@article{arxiv:2509.13638,
author = {{Qazalbash}, Meesum and {Zeeshan}, Muhammad and {O'Shaughnessy}, Richard},
title = {GWKokab: An Implementation to Identify the Properties of Multiple Population of Gravitational Wave Sources},
journal = {arXiv preprint arXiv:2509.13638},
year = {2025},
url = {https://arxiv.org/pdf/2509.13638v1}
}
@Misc{gwkokab2024github,
author = {{Qazalbash}, Meesum and {Zeeshan}, Muhammad and {O'Shaughnessy}, Richard},
title = {{GWKokab}: A JAX-based gravitational-wave population inference toolkit for parametric models},
url = {https://github.com/gwkokab/gwkokab},
year = {2024}
}