alt Hacker NewsThis outperforms Gemini 3 pro image (nano banana pro) on Text-to-Image Arena and Image Edit Arena. I'm surprised they didn't mention this leaderboard in the blog post.
I like this benchmark because its based upon user votes, so overfitting is not as easy (after all, if users prefer your result, you've won).
Replies
nycdatasci • today at 2:36 AM
The arena concept doesn’t work for image models due to watermarks.
➕ show 1 reply
The score are really, really close, it might be why