Reverse engineered prompt

GitHub

Build me a Windows desktop app in Python called QualityScaler that can upscale and clean up both images and videos with AI, all running locally on the user’s PC with no internet needed after setup. I want a polished, simple GUI that feels easy for normal people to use, where they can drop in photos or videos, pick an AI model, choose the GPU or CPU, resize before upscaling, and control how much of the original gets blended with the enhanced result.

It should support common image formats like jpg, png, webp, tif, bmp, and heic, plus common video formats like mp4, mkv, avi, mov, webm, gif, and others. Please include multi GPU support for AMD, Intel, and Nvidia on DirectX 12 hardware, automatic tiling for low VRAM cards, batch processing, video stop and resume, time remaining, audio kept in the final video, hardware accelerated encoding, and saved user settings.

Use local AI models from the project folders and ffmpeg from assets. Look up current docs online if you need to.

Have a live product UI? Try website reverse