loading
loading
The maximum amount of text a model can read and reason over in one go.
Everything the model can 'see' — your system prompt, conversation history, tool results, and documents — must fit inside this limit, measured in tokens. Exceeding it causes older content to be truncated or errors out entirely. Managing what goes in and out of the context window is one of the core engineering challenges when building agents.
Plainly
Think of Context Window as a named building block in a big LEGO app. The maximum amount of text a model can read and reason over in one go.
In practice
Use it when you are mapping how the app is structured or explaining a feature to a teammate or agent. In practice, define the owner, input, output, and failure mode before you rely on it.