Context Window
The maximum amount of text, measured in tokens, that a model can hold in mind at once, including your prompt and its own reply.
Go past it and the model forgets the earliest parts. It sets how much document or history you can work with in one shot.
Frequently asked questions
What is a context window in plain terms?
It is the maximum amount of text, measured in tokens, that a model can keep in mind at one time, counting both your input and its own reply. Think of it as the size of the AI's working desk: only so much fits on it at once.
Why does the AI seem to forget what I said earlier?
Because once a conversation or document runs past the context window, the earliest parts drop out of view and the model no longer sees them. Longer chats and big documents are where this shows up most.
How much can I paste into an AI at once?
That depends on the model's context window, which sets the ceiling on how much document or history fits in a single request. If you need to work with something bigger than the window, you break it into parts or use a technique like RAG to pull in only the relevant pieces.
New to all this? Start with what an AI agent really is, browse the full glossary, or explore the learning hub.
← Back to the glossary