Context window

The context window is the maximum amount of text, measured in tokens, that the language model can “see” at once and take into account when generating a response—including the entire conversation history, attached documents, and system instructions. A larger context window allows the model to work with longer documents, entire books, or extensive code bases without losing important information from the beginning of the text. If the prompt exceeds the size of the context window, the model will “forget” the older parts of the conversation.