Lightning

Lightning AI is a PyTorch-based platform that helps users painlessly train and deploy AI models between local machines and cloud environments. The framework provides a series of easy-to-use tools and interfaces, allowing users to train and optimize models more efficiently, and easily deploy trained models to production environments. It supports the construction of various popular AI models such as large language models, Transformers, Stable Diffusion, etc. Key features include support for distributed multi-GPU training, built-in...

The open source artificial intelligence platform Lightning AI announced a partnership with Nvidia to release a next-generation AI compiler called "Thunder" to accelerate the training of artificial intelligence models. According to Lightning AI, the compiler can increase the training speed of large language models (LLMs) by up to 40% compared to unoptimized code in actual scenarios. This good news brings new technological breakthroughs to the field of AI. The release of the "Thunder" compiler marks the first time that Lightning AI has developed a new compiler for the PyTorch open source machine learning framework...