⬢github Python · 356 ★ +38 since we first saw it · pushed 3 h ago · Apache-2.0
google-research/kauldron
Modular, scalable library to train ML models
Kauldron is a Python library from Google Research for training machine learning models, built on JAX and Flax. It's modular—datasets (TFDS, Grain), models, and optimizers plug in freely—and aimed at research velocity with ready-made example configs, Colab-first prototyping, integrated profiling, and runtime shape checking.
Why now: It gained attention from a Hacker News discussion showcasing the library, alongside its continued active development and recent citation-ready release.
Who it is for: ML researchers and engineers using JAX/Flax who want a configurable, composable training framework.
deep-learningjaxpython
Stars over our 287 snapshots: 318 to 356, since 2 d ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/google-research/kauldron