scope:

This standard defines a set of application programming interfaces (APIs) for use with different deep learning (DL) inference engines. The interfaces include parameter reading, model compilation optimization, operator registration, thread management, input/output data acquisition, inference instance creation, and inference execution.

Purpose

This standard defines efficient APIs for model management, device management, input/output, and inference that break down the barriers between different DL inference engines and applications. The target DL applications include natural language processing, image classification, and object detection.