What is a large language model (LLM)?
Also called: LLM, LLMs
A large language model (LLM) is an AI model trained on huge amounts of text to predict the next word. Doing that well at scale lets it write, summarise, translate, answer questions and write code. ChatGPT, Gemini and Claude are chat apps built on large language models.
How does a large language model work?
An LLM breaks text into tokens and, given everything so far, predicts the most likely next token, then the next, and so on. It learned these patterns from training on books, websites and code, and is usually fine-tuned afterwards to follow instructions helpfully.
Why does a large language model matter?
LLMs are behind most new AI tools. Knowing that they predict likely text, rather than look up facts, explains both their power and their mistakes.
An example
Example: Ask an LLM for a WhatsApp message wishing your team a happy Diwali, and it writes one word by word from patterns it learned.
Related terms
Learn it properly
A large language model is covered in AI Foundations, with short lessons, practice questions and instant feedback.