Large language model - Wikipedia

A large language model (LLM) is a neural network trained on extensive text data for natural language processing tasks like generation and translation, often based on transformer architecture which allows better parallelization compared to earlier models. LLMs have evolved from early statistical models in the 1990s through various advancements, culminating with the introduction of transformer-based models at NeurIPS 2017, leading to significant improvements in language understanding and generation capabilities.