AI Prompt EngineeringModuleLesson 01
What is an LLM?
15m
Beginner
+50 XP
0% Done
LLM stands for Large Language Model. It is a specific type of Generative AI designed explicitly to understand, generate, and interact using human language.
Models like ChatGPT (OpenAI), Claude (Anthropic), and Gemini (Google) are all LLMs. They are "Large" because they are trained on absolutely massive datasets—often encompassing a significant portion of the entire internet, including books, articles, code repositories, and Wikipedia.
Think of an LLM as a highly advanced version of the autocomplete on your smartphone.
When you type "I'm looking forward to...", your phone might suggest "seeing you". It does this because statistically, "seeing you" often follows that phrase.
An LLM does the exact same thing, but on a massive scale. Instead of predicting the next word based on a few recent texts, it predicts the next word based on the entire context of your prompt and the trillions of words it read during training.
You might hear people say "This model has 70 billion parameters." A parameter is essentially a "connection" or a piece of learned knowledge in the AI's neural network.
More parameters generally mean the model can grasp more complex concepts, nuances, and logic, but it also makes the model slower and more expensive to run.
Click a concept to explore its meaning.
A massive AI model designed specifically to process and generate natural language.
Why do we call these models 'Large' Language Models?
LLMs are the ultimate 'autocomplete'. They don't think; they calculate the statistical probability of what word should come next based on billions of internal parameters.