Support for configuring locally hosted LLM servers

[amithkoujalgi]

Hello.

First of all, thank you for creating this plugin. Its amazing! 🤩

I was wondering if we could have support for configuring locally hosted LLM servers in different ways. For instance, I can have an Ollama docker setup running locally with an exposed port. This way, it helps me isolate the LLM server and yet fully utilise it capabilities.

Let me know what you think. Thanks!

[uptopoint]

Hi All, If anyone has solution to this problem.
I want to setup a ollama base url as only one of my machines support running Ollama. Other machines have terrible performance issues.
But in codegpt I cant setup base url to some other machine and I cant use ollama.

Anyone has any pointers in that if there is a way to do this