AI MaxxingIntermediate75 sec

Context compression

Context compression reduces a long piece of text into a shorter representation that preserves the information a model needs, so it fits within a limited context window.

Mark as completed

Prerequisite: Context window

What matters

This lets a system handle documents far larger than the model's context window, at the cost of some detail loss compared to supplying the full original text.

Example

A 100-page report might be compressed into a 2-page summary that a model uses as context, instead of the model needing to read the full document at once.

Common misconception

Context compression does not guarantee no important detail is lost; a summary can accidentally drop information a later question depends on.

Recall question

What trade-off does context compression make to fit large documents into a context window?

Keywords and related concepts