perhaps the server part should use python instead of nodejs

[diemus]

I have investigate some python lib like langchain and gpt_index, it could create something like new bing or even more powerful function. perhaps the server part should be implement by python and use some libs like langchain to extend chatgpt abilities. I have found some interesting repo : https://github.com/circlestarzero/EX-chatGPT/blob/main/README.en.md

[cogentapps]

My current thinking about LangChain (and similar projects) is that if it's added to this project, it should be as part of an extension/plugin system. I'm picturing a screen in the app that shows a list of optional integrations the user can enable or disable.

However, I want the application architecture to be more stable and to have a better idea of what specific integrations are wanted before starting work on this.

For now, my goal is to keep the server lightweight and implement as much functionality as possible in the browser instead - so that we have the flexibility to make decisions like this (switching to Python) at a later date if needed.

[diemus]

it should be as part of an extension/plugin system. I'm picturing a screen in the app that shows a list of optional integrations the user can enable or disable.

yes, I think this is a direction for future llm.

For now, my goal is to keep the server lightweight and implement as much functionality as possible in the browser instead

I agree, this way i could put this web on a object store, without a backend. save a lot of money