The facial exam (emotions) was quite interesting, although not very accurate, I had to spend over 1 minute on some of the emotions. At least it's powered by a very tiny ML model that runs locally in the browser, powered by https://github.com/justadudewhohacks/face-api.js

> The face expression recognition model is lightweight, fast and provides reasonable accuracy. The model has a size of roughly 310kb and it employs depthwise separable convolutions and densely connected blocks. It has been trained on a variety of images from publicly available datasets as well as images scraped from the web. Note, that wearing glasses might decrease the accuracy of the prediction results.

what level is that?