Large language model
A large language model is a neural network trained on huge amounts of text to predict and generate language, the engine behind most AI chat and writing tools.
Category
Industry concept
See all terms
What is a large language model?
A large language model, or LLM, is a neural network trained on huge amounts of text to predict the next word in a sequence. That single task, repeated across billions of examples, teaches it grammar, facts, and patterns of reasoning well enough to write, summarize, and answer questions in plain language. Every chatbot, coding assistant, and AI agent people talk to runs on an LLM under the hood.
How is a large language model different from AI?
AI is the broad field of building systems that perform tasks normally requiring human intelligence, covering everything from image recognition to robotics. A large language model is one specific type of AI system, built for language and trained by predicting text rather than by rules a programmer wrote by hand.
How does a large language model work?
Training happens in two stages. Pretraining feeds the model a huge slice of text and adjusts billions of internal parameters so it gets better at predicting the next token. Fine-tuning then narrows that raw capability into something that follows instructions and holds a conversation, usually with human feedback shaping which responses count as good. The result is a transformer, a network that uses self-attention to weigh every token against every other token in its context window before producing the next one. None of this gives the model a way to look up a company's current knowledge. Qontext's context repository fills that gap, feeding the model the structured, permission-aware slice of company context it needs at request time.