Large Language Models (LLMs) are machine learning models that have been trained on large amounts of text data to understand and generate human language. These models are typically Transformers with massive number of parameters (100s of millions to 100s of billions).