LLM

Definition

A Large Language Model (LLM) is a deep learning model trained on massive text datasets to understand and generate human language. Modern LLMs can reason, code, summarize, translate, and generate creative content.

Simple Explanation

An LLM is fundamentally a very sophisticated autocomplete. It predicts the most likely next token (word, punctuation, or character) given all previous tokens. What makes it powerful is the scale — trillions of parameters and training data from books, websites, code, and more.

Real-World Use Case

You ask an LLM 'explain quantum computing like I'm five.' It generates a sequence of tokens that forms a coherent, simplified explanation.

Examples

Related Terms

Learn More

• OpenAI Documentation
• Anthropic Claude Docs