Hacker News new | ask | show | jobs
by paul_brook 829 days ago
Very cool! Good job guys. Is it possible to quantify the accuracy in any way?
1 comments
john_noumenonic 828 days ago
The preliminary model gets ~93% accuracy on an "in-the-wild" test dataset of ~10k images pulled from the widely used models, i.e. midjourney v5/v6, SDXL, DallE-3, etc.
link