01In plain English
A large language model is an AI model trained on a very large amount of text, which it uses to predict and generate language. GPT, Claude, Gemini and Llama are families of LLMs. Most AI writing, chat and coding tools are built on one of them.
02How it works
An LLM is trained by reading a huge collection of text and learning which words tend to follow which. When you type a question, the model does not look up an answer in a database. It breaks your text into pieces, then predicts the most likely next piece, adds it, and repeats until the reply is complete. Because it learned patterns of language, it can write, summarize, translate and explain. Think of the autocomplete on your phone, scaled up enormously and trained on far more material. A writing tool, a chatbot and a coding assistant may all sit on the same underlying model, with different instructions and an interface wrapped around it.
03Why it matters when you are choosing
Many tools are a layer on top of someone else's LLM. Knowing which model a tool uses tells you a lot about its quality, its limits and where your data goes.
04What to check
Check which model or models the tool uses, whether you can choose, and whether your data is passed to the model provider.
05Where you will meet it
06Common questions
What is the difference between an LLM and a chatbot?
An LLM is the engine, and a chatbot is the product built around it. The chatbot adds the chat window, memory, safety rules and sometimes extra tools. That is why two chatbots using the same model can behave differently, and why one model can power many products.
Does an LLM understand what it writes?
Not the way a person does. It predicts likely language from patterns it learned, which often produces correct and useful text, and sometimes confident mistakes. Treat its output as a strong first draft that needs checking, especially for facts, numbers, quotes and anything with legal or financial weight.
Do all AI tools use the same LLM?
No. Some use a model from a big lab such as OpenAI, Anthropic or Google, some use open models they host themselves, and some let you pick. Many tools do not say which model they use, so ask, since the model affects quality, speed and where your data goes.