fundamentals

Context Window

The maximum amount of tokens a model can consider in one request, including prompt and output.