G
GEO Toolbox

How AI Search Works

Context Window

Also: context length

A context window is how much text, measured in tokens, a language model can consider at once, including the prompt and any retrieved sources. In AI search, retrieved pages are loaded into the context window so the model can ground its answer in them.

Updated

When an AI engine retrieves sources, it fills the context window with their text and generates an answer from what fits. Content that is clear and chunked into self-contained passages is easier to place into that window and quote accurately.

The window is finite, so engines favor passages that answer the question directly over long preambles. That is part of why answer-first structure earns more citations.