What is LLM (large language model)?

An LLM is a neural network trained on vast amounts of text to predict the next token — the technology behind ChatGPT, Claude, Gemini and Grok.

A large language model learns statistical structure from text at enormous scale: grammar, facts, styles, reasoning patterns. Given some text, it predicts what comes next, one token at a time. Everything it does — answering, summarising, coding, translating — is that prediction repeated.

Chat products add a layer on top (instruction tuning, safety training, tools, memory), but the core stays a predictor. Understanding that explains both its strengths (fluency, breadth, pattern-matching) and its failure modes (hallucination, sensitivity to phrasing).