AI Systems

Inference, explained

Inference is the process of running a trained model on new inputs to produce outputs.
Back to Glossary

Part of the Semantic Notion technical glossary.

Short Definition

Inference is the process of running a trained model on new inputs to produce outputs.

Intuition

Training teaches the model. Inference uses the trained model in an application.

Technical Definition

Inference executes the learned model function with fixed parameters, often under latency, throughput, cost, and reliability constraints.

Example

A production API performs inference when it returns a fraud score for a transaction.

Common Misunderstandings

Inference is not training.
Inference quality depends on preprocessing, prompts, retrieval, and system design, not only model weights.

Start here

Need the broader concept map?

Return to the glossary index to move from this definition into adjacent AI, ML, mathematics, and computer science terms.