#Please increase number of tokens can return from custom actions' response in custom GPTS

1 messages · Page 1 of 1 (latest)

upper leaf
#

ChatGPT shared link: 40000 number of words are rejected
https://chat.openai.com/share/e/3266931b-b629-4330-a3c2-59a742f4f9e9

25000 worked https://chat.openai.com/share/e/2ef0cc79-af2b-437e-ada0-be357ffdfa66

Steps to reproduce:

  1. Return 40000 words from custom action

Expected result:
Model's input tokens increased to 4X from 32K to 128K We expect an increase in custom gpt response allowance

Actual result:

On our tests we could double the response keyword numbers but it's not obvious what exactly it should be

Additional information

Browser:
OS:

I think it will improve the experience greatly if you could allow more tokens to be returned from custom actions. If that's already the case, could we get more information about it? How many tokens can we send maximum to avoid getting the ResponseTooLarge error?

stray onyxBOT
#

@upper leaf

Please follow the posting guidelines.

For better assistance from the community or an OpenAI team member, please follow the posting guidelines by clicking the button below.