I've tried running a few open source apps like this locally, sometimes it's as simple as standing up a Mock OpenAI endpoint but if they use fancy features like

I've tried running a few open source apps like this locally, sometimes it's as simple as standing up a Mock OpenAI endpoint but if they use fancy features like embeddings endpoints then it can be more complicated. The quality is getting better, but it's always not-the-same.