onednn
Fast inference engine for Transformer models
High-performance Spiking Neural Networks Library Written From Scratch with C++ and Python Interfaces.