ONNXIFI with Facebook Extension

Intel® Optimization for Chainer*, a Chainer module providing numpy like API and DNN acceleration using MKL-DNN.

Collective communications library with various primitives for multi-machine training.

Tensors and Dynamic neural networks in Python with strong GPU acceleration

FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/

A tensor-aware point-to-point communication primitive for machine learning