Glossary
What Is a Large Language Model (LLM)? Definition and Example
A large language model, or LLM, is an AI system trained on huge amounts of text to understand and generate human language by predicting the most likely next words.
What it means
LLMs power chat assistants such as ChatGPT, Claude and Gemini. They break text into small pieces called tokens and use patterns learned in training to produce a reply. They can summarise, translate, draft and explain, but they do not look facts up unless they are connected to search or documents.
Example
When you paste a long report into an assistant and ask for the five main findings, an LLM reads the text and writes the summary.
Why it matters
Knowing that an LLM predicts text, rather than checking facts, explains both its strengths and why it sometimes makes things up.
Related terms
- Token (AI): A token is a small piece of text, often part of a word, that an AI model reads and writes. In English, one.
- Context Window: A context window is the amount of text an AI model can consider at once, including your prompt, any documents.
- AI Hallucination: An AI hallucination is a confident answer from an AI that is not true, such as a made-up fact, number, quote.
Learn this live, for free
Join the free 90-minute live masterclass, every Friday, 7 PM IST (India) or 7 PM ET (everywhere else). No tech background needed.
Register for the masterclass nowEducation only. This glossary is general information, not financial, legal or career advice.
Frequently asked questions
Is ChatGPT an LLM?
ChatGPT is an assistant built on large language models. The same is true of Claude and Gemini.
Does an LLM know the latest news?
Not by itself. It knows what it learned in training unless it can search the web or read documents you give it.
What is the difference between an LLM and an AI agent?
An LLM generates text. An agent uses an LLM plus tools and a loop to work toward a goal.