We are trying to use chat gpt model 3.5 turbo for chat completion rest APIs in our application. We are not sure about the maximum token limit (Request + Response) for this chat gpt model using chat completion API. Could anyone please let me know the maximum token limit using chat completion API.
Below is the chat completion API link:
it’s worth highlighting that the default alias gpt-3.5-turbo is about to get upgraded to gpt-3.5-turbo-0125 within a few days (Feb 16th) and will therefore have a 16k input window, so you may not have to update your model name.
Do you know if the 16k tokens of context window that they talk about for the gpt-3.5-turbo-0125 include the 4k tokens of output?
