Özer Cem KELAHMETKELAHMET
Glossary

What is Mixture-of-Experts? Definition and meaning

A layer with many expert networks where a router activates only a few per token, growing capacity without growing compute.

Definition

Mixture-of-Experts: A layer with many expert networks where a router activates only a few per token, growing capacity without growing compute.

In short: Route each input to a few specialist sub-networks.

When it is used

Scaling model capacity efficiently in large language and foundation models.

Learn more

The full guide to Mixture-of-Experts explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.

Previous term: Mixed precision · Next term: MLflow

Last updated Nov 18, 2026
Type to search.
    ↑↓ navigate↵ open