This is a high-performance and lightweight AI model inference engine specifically designed for Apple M series chips. It fully utilizes the characteristics of Apple hardware to enhance inference speed and provides a simple and easy-to-use API to help you deploy efficient local large model services with one click.