Large Language Model (LLM)

A large language model is an AI system trained on enormous quantities of text — books, articles, websites, code, conversations — to understand and generate human language. When you use ChatGPT, Claude, or Gemini, you are interacting with an LLM.

What makes them "large"

The "large" in large language model refers to two things: the size of the training data (often trillions of words) and the number of parameters in the model. Parameters are the internal settings the model adjusts during training to improve its predictions. Modern LLMs have hundreds of billions of parameters. More parameters generally means more nuanced understanding, but also more computational cost to run.

How LLMs generate text

LLMs work by predicting the next word (technically, the next token) in a sequence. When you type a prompt, the model calculates the probability of every possible next word, picks one, then repeats the process for the word after that, and so on. This happens incredibly fast — often generating hundreds of words per second.

This prediction mechanism is why LLMs can write essays, answer questions, summarise documents, translate languages, and write code. They have seen so many examples of each task during training that their predictions are remarkably useful.

What LLMs can and cannot do

LLMs excel at language tasks: writing, editing, summarising, translating, explaining, brainstorming, and analysing text. They can also handle structured reasoning, coding, and data analysis when given clear instructions.

However, LLMs have important limitations:

• They do not have access to real-time information unless connected to the internet or external tools. Their knowledge has a training cutoff date.
• They can hallucinate — generate plausible-sounding but incorrect information — because they are optimising for statistically likely text, not verified truth.
• They do not remember previous conversations unless the conversation is within the same context window.
• They struggle with precise mathematical computation, though this is improving rapidly.

The major LLMs you should know

• Claude (Anthropic) — Known for nuanced reasoning, long context windows, and careful handling of sensitive topics.
• GPT-4o/ChatGPT (OpenAI) — The most widely known LLM, strong across general tasks.
• Gemini (Google) — Integrated with Google's ecosystem, strong at search and multimodal tasks.
• Llama (Meta) — An open-source model that can be self-hosted, giving organisations more control over their data.

Each model has different strengths, pricing, and privacy characteristics. The right choice depends on your specific use case, budget, and data sensitivity requirements.