Large Language Model (LLM)

A large language model (LLM) is a neural network trained on vast amounts of text to predict the next token in a sequence. From that single objective emerges the ability to answer questions, write and summarize, translate, and drive tool-using agents. 'Large' refers to both the training data and the billions of parameters the model learns.

Because an LLM works by consuming and producing tokens, everything about running one - cost, latency, and rate limits - is measured in tokens. That property is what turns LLM usage into an economic and governance problem at enterprise scale.