← Dictionary
Dictionary · AI

What is Inference?

It is the process of a trained artificial intelligence model making predictions on new data.

Overview

It is the process of a trained artificial intelligence model producing an answer to a new question or input using the information it has learned. This stage is the moment when the model no longer 'learns' but 'applies'.

Analogy: It is the process of a chef, after years of learning to cook, going into the kitchen and cooking a meal for you from scratch.

How it works

The user asks a question, the model passes this question through its huge digital networks and calculates the most logical answer step by step and displays it on the screen.

Where it is used

The inference process occurs whenever you chat with ChatGPT or have an artificial intelligence draw a visual.

Commonly confused with

It is confused with training the model; The education model is the school years, while the inference is graduating and starting to work.

Frequently asked questions

How long does the inference process take?

It may vary from milliseconds to seconds, depending on the size of the model and the power of the computer.

Does the model learn anything new during Inference?

No, the model only makes predictions using the available information it has.

Related terms

Related tools

This explanation was written in plain language for TreScout and machine-translated from the Turkish original · the Turkish version prevails. If something looks wrong or missing, write to hello@trescout.com. Read in Turkish →