#tokens

Question 1: Single choice

A model has a context window of 128k tokens and can generate at most 16k output tokens. What is the context window?

The max_tokens setting of a request, which caps how long the reply can be1
The number of previous chat messages that the model takes into account when replying2
The most tokens the model can process at once, input and generated output together3
A setting that controls how creative the model's replies are4