mzguntalan/zephyr
Zephyr is a declarative neural network library on top of JAX allowing for easy and fast neural network designing, creation, and manipulation
This library helps machine learning engineers and researchers design and build neural networks quickly and efficiently using JAX. It takes a function describing the network's structure and automatically handles the complex parameter initialization, outputting a ready-to-use, JAX-compatible neural network function. This is ideal for those who prefer a purely functional approach to deep learning model development.
No commits in the last 6 months.
Use this if you are a machine learning engineer or researcher who works with JAX and wants to define neural networks as simple functions, minimizing boilerplate code and focusing on the network's mathematical structure.
Not ideal if you are accustomed to object-oriented neural network frameworks (like PyTorch or TensorFlow Keras) or if you are not already working within the JAX ecosystem.
Stars
36
Forks
1
Language
Python
License
Apache-2.0
Category
Last pushed
Sep 11, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mzguntalan/zephyr"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
explosion/thinc
🔮 A refreshing functional take on deep learning, compatible with your favorite libraries
google-deepmind/optax
Optax is a gradient processing and optimization library for JAX.
patrick-kidger/diffrax
Numerical differential equation solvers in JAX. Autodifferentiable and GPU-capable....
google/grain
Library for reading and processing ML training data.
patrick-kidger/equinox
Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/