AcuiRT

AcuiRT (Advanced Conversion UtIlities for Run-Time) is a model auto‑conversion library for fast inference of AI models using device‑specific deep learning compilers. By using AcuiRT, you can perform high‑speed model inference on actual hardware with deep learning compilers such as TensorRT.

Features

AcuiRT provides an automatic conversion feature integrated with a deep learning compiler, automatically converting models into efficient ones. By using the converted model, fast inference can be achieved.

Users can focus on designing models. When compiling a model that includes layers that cannot be compiled, existing tools required manual module partitioning and trial-and-error, but AcuiRT automatically partitions modules and determines compile feasibility, reducing the user's effort.

Quick Start Guide

• Setup
• Basic usage
• Tutorial

Conversion Tips

AcuiRT also supports manual module conversion in addition to automatic conversion.

• Module-specific conversion