MegEngine

MegEngine is a free deep learning framework for developing and deploying models, with automatic differentiation and local training. It uses one model for training and inference, including quantization and dynamic shapes. The project says enabling DTR can reduce GPU memory use to one-third of the original. Inference support spans x86, Arm, CUDA, and ROCm, while GPU use requires compatible device drivers. Python packages are listed for 64-bit Linux and Windows, macOS 10.14 or later, and Android 7 or later; macOS and Android installations are CPU-only. The installation guide lists Python 3.6–3.9. For deployment, MegEngine Lite provides C/C++, Rust, and Python runtimes. Related tools include MgeConvert for converting between MegEngine and third-party model formats, MegFile for S3, HTTP, and local file interfaces, and MegFlow for streaming computation in AI applications. The project presents tutorials for beginners and advanced developers, and lists GitHub issues, a forum, a QQ group, and email for support.