Large language model (LLM): definition

A large language model is a neural network trained on huge amounts of text to predict the next token, which lets it write, summarise, classify, and answer questions across almost any topic.

Models like GPT, Claude, and Gemini are large language models. Under the hood they do one thing, predict the most likely next piece of text, but at scale that single skill turns into drafting, translation, code, and reasoning.

For a business the practical point is that an LLM is a general engine you steer, not a finished product. The value is in how you feed it (your data, your rules) and what you wire it to (your tools), which is the part we build.

Related terms

AI agent: An AI agent is a system that uses a language model to decide which steps to take and which tools to call to finish a goal, instead of just answering a single prompt.
Inference: Inference is the act of running a trained model to get an output, and inference cost is what you pay each time the model answers, measured in tokens.
Fine-tuning: Fine-tuning is further training a base model on your own examples so it learns a specific style, format, or task more reliably than prompting alone can achieve.
Prompt engineering: Prompt engineering is the craft of writing the instructions, examples, and context you give a model so it produces reliable, correctly formatted output for a specific task.