nagadomi/nunif — reverse-engineered prompt

I want a local AI media toolbox that wraps the main stuff in this repo into something I can actually use without digging through a bunch of scripts. The biggest things I care about are waifu2x for making anime art and photos bigger and cleaner, iw3 for turning normal 2D images and videos into side by side 3D for VR, including a simple way to view or stream the result to a headset, stlizer for basic video stabilization, and cliqa so I can score or filter low quality images when I am preparing a dataset.

Please make it feel practical and polished, with sensible defaults, pretrained models handled for me, and clear install and run steps. Windows matters most, but if the repo already supports macOS and Ubuntu, keep that in mind too. If there is already a desktop app or player in here, hook that up instead of rebuilding everything from scratch. I want it to run locally with Python, and use GPU acceleration when available. Look up the current docs online if you need to.