LLM
Large language model: a program trained on an enormous pile of text to predict the next word. ChatGPT, Claude, and Gemini are all LLMs wearing different outfits.
An LLM (large language model) is a machine learning model trained on a huge amount of text (books, websites, code, forum arguments) to do one thing: given some text, predict the next token . Do that billions of times during training and the model ends up holding a compressed picture of how language, and a lot of the world described in language, tends to go. Nearly all modern LLMs are transformers , most of them in the GPT style.
The “large” is about parameters : the learned numbers inside, now counted in billions. Size buys ability, but it also buys cost: more memory to run it, more electricity to train it. ChatGPT, Claude, Gemini, Llama, and Qwen are all LLMs; the differences are in training data, size, and the fine-tuning that turns a raw text-predictor into something that answers questions politely.
The thing worth keeping in your head: an LLM produces plausible text, and plausible usually means true, but not always. That gap is where hallucination lives.