Parameters

Also known as: Model Parameters, Weights

In one sentence

Numbers inside an AI model that get adjusted during training to improve accuracy. More parameters usually mean more capability.

Explain like I'm 12

Think of parameters like knobs on a big mixing board. During training, the AI tweaks millions of knobs until the output sounds right.

In context

A model with 175 billion parameters has 175 billion numbers that control its behavior. GPT-3 has 175B parameters, GPT-4 has even more.

Related Guides

Learn more about Parameters in these guides:

Structured Output and Function Calling: Getting Reliable JSON from AI

Intermediate

Learn how to get reliable, parseable JSON output from AI models using structured output, function calling, and JSON schema. Essential for production AI applications.

15 min read

Temperature and Sampling: Controlling AI Creativity

Intermediate

Temperature, top-p, and other sampling parameters control how creative or deterministic AI outputs are. Learn how to tune them.

6 min read

What is AI? A Friendly Primer

Beginner

A non-jargony intro to AI, machine learning, and large language models. Learn the fundamentals without getting lost in technical details.

7 min read