Context window
Also known as: context length
A context window is the maximum amount of text (measured in tokens) an AI model can take into account in a single conversation.
In more detail
Everything in the current exchange — your instructions, any documents you paste, the back-and-forth so far, and the model's answer — has to fit inside the context window. When a conversation gets very long or you paste a huge document, older material can fall outside the window and effectively be forgotten. Different models have different context-window sizes, and they change over time, so check the tool's own documentation for the current figure.
Example
If you paste a 200-page report that exceeds the model's context window, it can only reason about the part that fits. Summarizing the report in sections is a common workaround.
Why it matters
Knowing the context window exists explains why an AI sometimes 'forgets' earlier instructions in a long chat and why very large documents need to be split up.
Prompts that use this
Sources
- 1.Large language model — Wikipedia (context window)
Wikipedia · verified 2026-07-26
Supports: A model's context window bounds how much text it can process at once.
Try a prompt for yourself
Open any prompt and run it in the IterPrompt playground.