How to store the instance of the gpt model

I am developing a NLP based feature exploration application using GPT-3. In this application, i am adding examples for how to read a csv into data frame, how to list the unique values of a column in a data frame, how to do outlier detection. The idea is to have data scientists and engineer use plain english to do feature exploration.
I am able to add examples to the gpt model, but I wanted to know once I add the examples how do I store this model and retrieve this through another UI based application where users will provide the prompts and the output of the code will be displayed. Right now my prompts and the examples are in the same python code. I want to use two different python program, one to add example and the other to input prompt and get the response.