GPT Builder Quality of Life Improvments

Greetings
I am trying to use chat GPT Build. There is not enough info or tools to allow for a proper development for a non-coder. It needs some simple quality of life improvements to make this work better.

  1. What does the 20 file upload limit mean?
  2. Where is the upload history of files?
  3. We need to know which files have been assimilated into the bot which ones have not, why they are having issues?
  4. Allow us to include links to some responses?
  5. Create a library of uploaded images, files or Dalle creations - more organization.
  6. Why does Code Interpreter need to be enabled to upload files?
  7. Some basic guides on building would be useful.

Thanks.

  1. You can upload up to 20 files when building the GPT.

2-3.

Look at the “Knowledge” section.

  1. You can tell your GPT to include links from your documents. The best way to do this it to try some prompt engineering.

  2. Not sure if you’re referring to using the GPT or creating it.

  3. Code interpreter is what allows the GPT to access files and read/manipulate them. Without it the model has no way of extracting the information from documents.

  4. Creating a GPT | OpenAI Help Center, the best way to get good at it is practice! Just try things and find what works best. There are lots of online guides for prompt engineering if you’d like to speed things up.

These three articles may help get started:

https://help.openai.com/en/articles/8554407-gpts-faq
https://platform.openai.com/docs/actions/introduction
https://suefel.com/evolve

1 Like