KANX: Production-grade Kolmogorov-Arnold Networks library (PyTorch + TensorFlow + ONNX)

[Min_Htet_Myet]

KANX is a production-ready implementation of Kolmogorov-Arnold Networks.

Key features:

• Native PyTorch and TensorFlow backends + ONNX export
• Docker and Kubernetes deployment support
• On smooth functions, smaller KAN models can outperform larger MLPs (with benchmarks)
• Over 92% test coverage

Development experience and lessons learned:

https://medium.com/@mattral-lifelong-learning/i-rebuilt-kan-networks-for-production-what-i-learned-391fd55914e0

GitHub: GitHub - Mattral/KANX: One library, four surfaces. Production-grade Kolmogorov-Arnold Networks || TensorFlow + PyTorch + ONNX. || A small KAN beats a 10× larger MLP on smooth, separable target. One library. Two backends. Real ONNX export. Docker + Kubernetes ready. · GitHub

Open source project. Feedback welcome.