Glossary
What is a context window?
The amount of text an AI can hold in view at once — your whole conversation so far. It re-reads this on every reply and has no memory beyond it; when it fills up, the oldest part drops off.
The context window is everything the model can see at once — your prompt plus the whole conversation so far, measured in tokens. The model re-reads all of it on every reply and has no memory outside it. When the window fills up, the oldest turns fall off the edge.
One fixed token budget
As the chat grows, the oldest turns drop off the start — the model never sees them again.
This is why a long chat starts "forgetting" the beginning, and why pasting a huge document can crowd out your actual question. Bigger windows help but aren't free — they cost more and can dilute focus, so what you put in the window is a real design choice.
Related terms
Appears in these lessons
See it live
Watch the window fill and forget
See what fits, what drops, and why — free in the app.