Iro AIIro AI

Iro AI Blog

What is a context window in AI?

A context window is how much text an AI can “see” at once — your prompt plus everything earlier in the conversation. Here's what it means in plain English and why it matters.

By ~5 min readAI Fluency

What is a context window in AI?
In this postDefinitionHow it worksWhy AI 'forgets'Practical tips

What is a context window?

A context window is the maximum amount of text an AI model can consider at once — your current prompt, any documents you've pasted, the earlier conversation, and the model's own reply, all together. Think of it as the model's short-term memory or its "field of view." It's measured in tokens (roughly, chunks of words), and every model has a limit.

How the context window works

Everything you send shares one budget. Your instructions, a long PDF you pasted, and the whole back-and-forth so far all count against the window. The model reads all of it to generate a response, and its answer counts too. A larger window means the model can take in more — a long report, a big codebase, hours of conversation — without losing track. Smaller windows fill up faster.

Practice this, don't just read it.

Iro AI turns ideas like the ones in this post into 5-minute exercises with feedback. Free tier, Pro from $5/month ($59.99/year, 7-day free trial).

Download Iro AITake the AI IQ test

Why AI seems to “forget” things

When a conversation grows past the window, the oldest content falls out of view. That's why a long chat can start contradicting itself, dropping details you mentioned early on, or "forgetting" instructions from the top. It isn't being careless — that text is simply no longer in its field of view. This is also why pasting a giant document and then asking many follow-ups can degrade: the document is crowding the window. Related: how to spot when AI is making things up.

Practical tips for working within the window

  • Put the important stuff up front and restate it if a chat gets long.
  • Start a fresh chat for a new task instead of letting one thread sprawl.
  • Paste only the relevant excerpt, not an entire document, when you can.
  • Summarize a long conversation and carry the summary into a new one.

Managing context well is a quiet superpower — part of broader AI fluency. You can build these habits in 5 minutes a day.

Practice this, don't just read it.

Iro AI turns ideas like the ones in this post into 5-minute exercises with feedback. Free tier, Pro from $5/month ($59.99/year, 7-day free trial).

Download Iro AITake the AI IQ test

Read next

FAQ

What is a context window in simple terms?

It's how much text an AI can pay attention to at once — your prompt, any pasted documents, the earlier conversation, and the model's reply, all together. Think of it as the model's short-term memory, measured in tokens.

Why does AI forget what I said earlier?

Because long conversations eventually exceed the context window, and the oldest content drops out of the model's view. It isn't being careless; that text is simply no longer something it can see, so details and early instructions get lost.

Is a bigger context window always better?

It helps for long documents and long chats, but it isn't everything. A clear, well-organized prompt that puts the important information up front often matters more than raw window size — and huge inputs can still dilute the model's focus.

How is a context window related to tokens?

Context windows are measured in tokens, the small chunks of text models process. The window is the maximum number of tokens — input plus output — the model can handle at once, so longer text uses more of it.

About the author

  • Alex Furukawa is the founder of Iro AI, the gamified app for learning to use AI well. He writes about practical AI fluency — prompting, AI tools, and the daily habits that turn AI from a novelty into real leverage.