A Large Language Model is a neural network — usually with hundreds of billions of parameters — trained on trillions of tokens (words, code, math, conversation) scraped from the internet, books, and licensed sources. Once trained, it can answer questions, write code, summarize documents, translate languages, and generate creative content. The 'Large' part matters: smaller language models existed for decades (autocorrect, predictive text), but it wasn't until OpenAI's GPT-3 (2020) that scale crossed a threshold where the models could handle open-ended reasoning. The dominant LLMs of 2026 — GPT-5, Claude 4.x, Gemini 3 — share the same underlying transformer architecture but differ on training data, alignment approach, and inference cost. LLMs do not 'understand' in the human sense — they predict statistically likely next tokens. This is why they sometimes confidently produce plausible-sounding nonsense (see: hallucination).
GLOSSARY
What is LLM (Large Language Model)?
An AI system trained on massive text datasets to predict and generate human-like text — the technology behind ChatGPT, Claude, Gemini, and most modern AI chatbots.
RELATED TERMS
Prompt Engineering
The practice of crafting input text (prompts) to get the best output from an LLM. A vague prompt gives vague answers; a structured prompt gives precise ones.
Transformer
The neural network architecture underlying modern LLMs and most image AI — introduced by Google in 2017 and quickly the dominant approach.
Hallucination
When an LLM confidently produces information that's factually wrong — invented citations, fake quotes, plausible-sounding but incorrect claims.
Context Window
The maximum number of tokens an LLM can process in one interaction — including your prompt, conversation history, and the model's response.
Token
The basic unit that LLMs read and produce. Roughly 0.75 words in English. APIs charge per token consumed and produced.
Back to the AI Glossary