You could compare it to the concept of security by obscurity which is obviously bad.
The difference to OpenWebUI is that you don't have to host anything and that it's native, compiled code written in flutter. This has multiple advantages:
1. We'll be able to ship the same features across platforms while still feeling native. (Android, iOS and macOS releases coming soon)
2. You don't have to run a docker container on your pc if you don't have a server
The model selection on OpenWebUI also shows every possible model without having the ability to choose which models you want to see in the selector. We solved that in AnyLM with custom model selection (it's pretty easy to understand if you try out the app).
Not to bash OpenWebUI. Honestly they did a very good job and it's even open source but yeah
About the linux support: You were actually the first one to ask me for this and it's definitely possible to do. I will put this in the roadmap :)
Hey HN, I'm excited to share AnyLM, a desktop app I've been working on. I love tinkering with local LLMs (via Ollama and LMStudio) but often found myself wishing for a more polished, unified UI similar to what ChatGPT offers, where all my conversations are in one place.Out of that frustration I've built AnyLM, which aims to solve that by providing a single, native application (currently for...