Skip to content

Pick your style: Denkbaar

Book a conversation

Articles

Knowledge

What is a large language model (LLM)?

A language model is an AI model that reads and writes text, and does not give the same answer every time.

A person compares two response cards beside a laptop.
A language model reads and writes text with varying answers. Illustration: Denkbaar.

A large language model (LLM) is an AI model that reads and writes text. It does not give the same answer every time.

What you read about it

  • For a piece of text, a varying answer is sometimes fine. For a quote, an invoice or a deploy it is not: an agent that does what you want.
  • In an AI pipeline the model reads text and turns it into fields. Rules and code check the fields: AI pipelines.
  • An agent is a language model that calls tools: What is an agent?.
  • A small open language model such as Phi-3-mini, with 3.8 billion parameters, runs on a laptop: Phi-3.
A document and a chip sit beside organised input fields.
Let code check the fields the model produces. Illustration: Denkbaar.

Why it matters to you

Use a language model to read and write, and let software calculate and a person decide. A small model suits fixed, simple tasks. For hard questions or long reasoning a large model is still better. So always test on your own examples before you choose.

A person selects a small chip beside a laptop.
Test models on your own examples before choosing. Illustration: Denkbaar.

Written on 13 November 2026, based on the articles and sources below.