oneDNN is an open-source, cross-platform library of optimized deep learning operations for CPUs and GPUs. It provides highly optimized implementations of matrix multiplication, convolution, attention, normalization, and other operations, and is the acceleration layer behind PyTorch, TensorFlow, OpenVINO, and many other AI frameworks.

oneDNN is intended for deep learning applications and framework developers interested in improving application performance on CPUs and GPUs. Deep learning practitioners should use one of the applications enabled with oneDNN.

This package is built in multiple configurations. To select a specific configuration install one of onednn-cpu-omp, onednn-cpu-tbb, onednn-cpu-threadpool, or onednn-dpcpp packages.

For more information visit oneDNN Github: https://github.com/uxlfoundation/oneDNN