Jim 🚬

A JAX-based gravitational-wave inference toolkit

Jim is a JAX-based toolkit for Bayesian parameter estimation of gravitational-wave sources. It pairs differentiable waveform models from ripple with GPU-accelerated JAX-based samplers, enabling massively parallel inference.

Supported samplers:

• flowMC — normalizing-flow-enhanced MCMC with optional parallel tempering.
• BlackJAX NS AW — nested sampling described in Prathaban et al. 2025.
• BlackJAX NSS — nested slice sampling described in Yallup et al. 2025.
• BlackJAX SwiG — nested slice sampling with waveform-cache reuse across parameter blocks described in Yallup et al. 2026.
• BlackJAX SMC — sequential Monte Carlo with optional adaptive tempering and persistent sampling.

For a quick introduction, see the Quick Start guide.

Installation

The simplest way to install Jim is through pip:

pip install JimGW

This will install the latest stable release and its dependencies. Jim is built on JAX. By default, this installs the CPU version of JAX. If you have an NVIDIA GPU, install the CUDA-enabled version:

pip install JimGW[cuda]

If you want to install the latest version of Jim, you can clone this repo and install it locally:

git clone https://github.com/GW-JAX-Team/Jim.git
cd Jim
pip install -e .

We recommend using uv to manage your Python environment. After cloning the repository, run uv sync to create a virtual environment with all dependencies installed.

Origins

Jim was originally developed as kazewong/jim by Kaze W. K. Wong and others. The original repository is no longer actively maintained; this fork is the active continuation of the project.

Attribution

If you use Jim in your research, please cite the accompanying paper:

@article{Wong:2023lgb,
    author = "Wong, Kaze W. K. and Isi, Maximiliano and Edwards, Thomas D. P.",
    title = "{Fast Gravitational-wave Parameter Estimation without Compromises}",
    eprint = "2302.05333",
    archivePrefix = "arXiv",
    primaryClass = "astro-ph.IM",
    doi = "10.3847/1538-4357/acf5cd",
    journal = "Astrophys. J.",
    volume = "958",
    number = "2",
    pages = "129",
    year = "2023"
}