JAXtronomy curates high-performance astronomy tools powered by JAX, bringing GPU-accelerated computing and automatic differentiation to cosmic research from gravitational waves to exoplanets.

Modern astronomy generates data at staggering scales—from LIGO’s gravitational wave detections to upcoming surveys that will catalog billions of galaxies. Traditional computational approaches often buckle under these demands, creating a bottleneck between observation and discovery. JAXtronomy addresses this challenge by curating cutting-edge astronomical software built on Google’s JAX framework, bringing the power of automatic differentiation and GPU acceleration to cosmic research.

This carefully curated repository spans the full spectrum of astrophysical phenomena, from ripple’s differentiable gravitational waveform modeling to flowMC’s normalizing-flow enhanced sampling for Bayesian inference. The collection includes tools for computational fluid dynamics simulating stellar explosions, exoplanet detection algorithms, cosmological parameter estimation, and even differentiable magnetohydrodynamics. Each tool leverages JAX’s NumPy-like API while delivering orders-of-magnitude speedups on GPUs and TPUs, making previously intractable problems computationally feasible.

Researchers worldwide are already using these tools to push the boundaries of discovery—from improving gravitational wave parameter estimation to modeling the complex physics of neutron star mergers. As astronomical datasets grow exponentially and physics simulations become increasingly sophisticated, JAXtronomy represents the future of computational astrophysics, where the universe’s most complex phenomena become accessible through elegant, high-performance code.

⭐ Stars: 52
💻 Language: Unknown
🔗 Repository: JAXtronomy/awesome-JAXtronomy