WaifuXL is our state of the art super resolution model trained on ~1,200,000 anime style images from the Danbooru2021 dataset. You can run it straight from your browser, entirely client-side, at https://waifuxl.com.

Using the Onnx Runtime, we stream the weights of our ML models directly onto your device to be executed locally in WebAssembly. Doing so allows us to provide this service solely through a static webpage, no backend for model execution needed. This has the added benefit of enabling the privacy of your images--your images are not, and never will be, sent to us.

Our original inspiration for this project was waifu2x (http://waifu2x.udp.jp). In general, our model will outperform those used on waifu2x by a significant margin, without the need for multiple models trained on various noise reduction levels.

Note that while you _can_ upscale natural (real) images, the model was only trained on anime style drawings, so don’t expect good results. It does work quite well on non-anime graphic art, however.

For more information, see our blog post at https://haydn.fgl.dev/posts/the-launch-of-waifuxl/