Growthtrait
All terms
Large Language Model

LLM

An LLM (Large Language Model) is an AI model trained on vast amounts of text to understand and generate human language. It works by predicting the most likely next piece of text given everything so far, which lets it answer questions, write, summarize, and hold a conversation.

Large language models are the engines behind most text-based AI, including ChatGPT, Claude, and Gemini. They shifted AI from narrow, task-specific tools to general systems that can handle a wide range of language work from a single model, which is why they now sit under so many products.

How Does an LLM Work?

An LLM breaks text into small units called tokens and learns, from enormous training data, the statistical relationships between them. To respond, it predicts the next token over and over, each time considering everything written so far, until a full answer is formed. It is not looking up facts; it is generating the most plausible continuation based on patterns it learned, which explains both its fluency and its failure modes.

What Is the Difference Between an LLM and a Foundation Model?

An LLM is the most common type of foundation model, but the terms are not identical. Foundation model is the broader category, including models that work with images, audio, and video, not only text. An LLM is a foundation model specialized in language. In everyday use the terms are often blurred, because the foundation models most people interact with are language models.

What Are LLMs Used For?

  • Answering questions and explaining concepts.
  • Writing and editing text, from emails to long documents.
  • Summarizing and extracting information from source material.
  • Powering chat assistants and, when connected to tools, AI agents.

What Are the Limitations of LLMs?

Because an LLM predicts plausible text rather than retrieving verified facts, it can generate confident but false statements, an AI hallucination. Its knowledge is frozen at a training cutoff unless paired with retrieval such as RAG, and it can reflect biases in its training data. These limits are why grounding and human review matter for reliable use.

Frequently asked questions

What does LLM stand for?+

LLM stands for large language model. It is an AI model trained on vast amounts of text that understands and generates language by predicting the most likely next piece of text, which lets it answer, write, summarize, and converse.

Is an LLM the same as a foundation model?+

Not exactly. An LLM is a foundation model specialized in language. Foundation model is the broader category and includes models for images, audio, and video. The terms are often used interchangeably because the best-known foundation models are language models.

What are examples of LLMs?+

GPT (behind ChatGPT), Claude, and Gemini are large language models. They power chat assistants, writing and summarizing tools, and, when connected to external tools, AI agents.

Related terms

Foundation Model

Foundation Model

A foundation model is a large AI model trained on a broad, massive dataset that can be adapted to many different downstream tasks. Rather than being built for one job, it is a general-purpose base that is then fine-tuned or prompted for specific uses. GPT, Claude, and Gemini are foundation models.

Generative AI

Generative AI

Generative AI is a category of artificial intelligence that creates new content, such as text, images, audio, video, or code, by learning patterns from large amounts of training data. Unlike systems that only classify or predict, generative models produce original output in response to a prompt.

RAG

Retrieval-Augmented Generation

RAG (Retrieval-Augmented Generation) is a technique that lets a large language model retrieve relevant information from an external source before generating an answer, rather than relying only on what it learned during training. It grounds the model's response in specific, up-to-date documents.