Context window

A context window is how much text an AI model can take in at once, measured in tokens: the question, the chat so far and any pages it reads.

AI searchUpdated 1 min read

Also called context length · context limit

What it means

Everything the AI considers when answering has to fit in its context window: your question, earlier messages, its instructions, and the text of any pages it read. Modern models can take hundreds of thousands of tokens, some a million.

Why it still matters

Big windows don't mean AI reads every page in full. AI search tools often read many pages and keep only the most relevant passages from each, to stay fast and cheap. See content chunking.

And models pay less attention to some parts of a long context than others. A fact buried in the middle of a long page is easier to miss.

What to do

  • Lead with the answer. Key facts first.
  • Use clear headings so the right section is easy to pick out.
  • Keep each section self-contained.

Example

An AI reads 15 pages to answer a question about project tools. From each, it keeps a few hundred words. A competitor's page has a clear "Pricing" section with a table, which gets picked. Your page mentions prices in passing across three paragraphs, and none of them make the cut.

Free tool: AI Page InspectorSee any page the way an AI crawler does: the text it gets, the passages it can quote, and checks with fixes.Try it →

Frequently asked questions

Does a bigger context window help my brand?

Indirectly. AI can read more pages, but you still need your facts in the parts it keeps.

What happens when a chat gets too long?

Older parts get dropped or summarised. That's why long chats can lose track of earlier details.

Is the context window the same as memory?

No. Memory features save notes between chats. The context window is what the model sees right now.

Go deeper

All posts →

Make AI search your next revenue channel

Start free trial