I just wanted to see people's opinions on using local on device LLM's. I am experimented with some, and while the simplest solution right now is to Nvidia Chat if you have an Nvidia GPU, I can see this space expanding a lot, especially with modern devices with NPU's