Feature Request: GUI for Multi-Model Selection and Dynamic AI Tuning

[FromLinuxWithLove]

I have recently started using Why AI and I would greatly appreciate it if a graphical user interface (GUI) were implemented that allows users to choose between multiple AI models. Currently, it appears only one model (Gemma4 E2B) is available with a standard configuration (e.g., 8192 tokens and a 32768 context window). Source: Why AI

Furthermore, I propose the addition of an option to adjust AI parameters, such as token limits and context window, dynamically via a simple button click, based on the specific system performance. This would allow the chosen AI model to automatically determine the most suitable settings for the underlying hardware.

For example, with my setup (RX 7700XT, 32GB RAM, Ryzen 5600X CPU), I could run significantly more powerful AI models than the pre-installed standard model allows.

I suggest adding Gemma 4 E4B, Gemma 4 12B alongside Gemma 4 E2B. It would also be highly beneficial if the user receives information about the specific system requirements, storage size and the target device class (e.g., Laptop, High-End PC, Office PC) before making a model selection.

[Anduin2017]

There is a huge blueprint for the AnduinOS AI Stack. Ultimately, with they why AI, user may able to communicate with the system with natural language. For example: What's wrong with my driver. And the why AI just start investigation.

[FromLinuxWithLove]

This sounds a bit like OpenClaw and I totally do NOT want an AI assistant controlling my system.

[Anduin2017]

Not controlling. Consider it like a claude code cli with a local model.