A large language model is a system trained on a very large body of text until it can predict, and therefore produce, language. Ask it to structure a draft, summarize a transcript, classify a lead, or draft a reply, and it does, in plain language.
Several labs ship them. They improve on a cycle of months. They are priced per use, and smaller ones can run on a desktop machine.
For our purposes the important fact is that they are interchangeable components. An agent's instructions do not care which one is answering. The best one for the job today may not be the best one next quarter, and switching does not require rewriting anything you own.
Example
What it looks like.
llm.mdplain text
One model reads a five-page legal draft and returns it structured under eight headings with the author's every statute number untouched. Ten seconds.Related
Read next.
- modelThe intelligence an agent calls when it needs to read, write, or decide. Replaceable. When a better one ships, the same instructions run on it and the results improve.Read the definition
- generative AIModels that produce new text, images, audio, or video from a description. The reason creative production now runs inside the same loop as everything else.Read the definition
- inferenceRunning a model to get an answer. It costs money per call in the cloud, or electricity on your own machine. Where it runs is a choice.Read the definition
- tokenThe unit a model reads and writes in. Roughly three quarters of a word. Models are priced and limited by tokens, not pages.Read the definition