I finally have “App” containers working with my Blackwell GPU in my home lab. I’d like to run a local LLM server that LM Studio, running on my Mac can connect to. Presumably this is via a running container with a REST API. Can anyone point me in the right direction to get started? I’m not sure if there are any existing apps in the catalog that can accomplish this. I had no trouble getting OpenWebUI to run a local model on the GPU today (yay… finally). Obviously I can just use OpenWebUI, or other app in a web interface. I’m just interested in how difficult or useful it would be to run a server.