Douglas Mourray

If whoever is responsible for this actually considers implementing any of my suggestions I will gladly help to come up with a sketch on how it could be done. I am a complete novice at this AI stuff but I work with Art Direction and UI design so I might be able to be of assistance.

The stuff people do here at hugging face is amazing. There are many here with fantastic knowledge about AI. There are also many here that probably have a decent computer knowledge in general but don't have any particular knowledge about AI, but that are interested in experimenting and learning more by running models locally. I am one of those. I am running stable diffusion and flux models with forge webui and some LLMs with LM studio. I have also tried coding with tools like Cursor and Windsurf. I can manage to get these models up and running with the help of lots of googling and YouTube tutorials. But I am not able to understand which of the models on huggingface that would work on my hardware limitations by just looking at the list of models available on hugging face. The nomenclature of the models is confusing to me.

Please please add a function so I easily can sort models by:

PRIORITY 1

HARDWARE REQUIREMENTS
Let me input my hardware specs and OS like; Windows 11, 32GB RAM, Nvidia 3070 8GB VRAM, Ryzen 5800x

WHAT INTERFACE THEY WORK WITH. (most people want to use an interface and not work in the terminal)
"LM studio" "forge web ui" and so on.

WHAT IS THE MODEL GOOD AT
like "image generation" "coding" "general purpose LLM"

PRIORITY 2

And as a bonus it would be great if it was made clear in as easy terms as possible

WHAT INPUT can the model take?
Text, code, images, videos, url?

WHAT OUTPUT does the model create?
Text, code, images videos and so on.

IS THE MODEL what I call "ACTIVE" or "PASSIVE" (I dont know the correct term for this. What I am referring to might be what is called "Agent".

In any case what I mean by this is if the model will only spit out results in a chat window for the user to copy or use by them self or if its possible for the model to create files, run terminal commands, access files in a certain folder and so on.