TUTORIAL

AI Terminology #21: Perplexity

A metric used to evaluate language models, measuring how well a model predicts a sequence. Lower perplexity indicates better performance.

Team Command Code

1 min read

Mar 23, 2026

A metric used to evaluate language models, measuring how well a model predicts a sequence. Lower perplexity indicates better performance.

Try Command Code

The best coding agent for open models, in your terminal or on your desktop.

1npm i -g command-code

Share this article

Coding taste, patterns, and strategies that actually work

ACP is now live in Command Code. Run cmd acp to use it in an ACP-compatible editor like Zed.

Mistral Large 4 is now live in Command Code. 1T total params. 49B active params. 524K context.

Introducing Agr, an open decision model by Command Code. 58.15 on Decision Index 0.2.1. Best-in-class on coding agent tool calls and routing.