Hacker News Reader

"Vision is an especially large weakness."

But you can have GPT write code to reliably convert the image grid into a textual representation, right? And code to convert back to image and auto-verify.

View on HN

GPT-4o might not have been trained on a sufficiently large amount of visual data to develop advanced spatial intelligence. Perhaps it needs to see a lot more images, or perhaps it needs to be trained differently (e.g. predict the next frame in a video). I suspect SORA has more spatial intelligence internally than 4o.

The task provides json grid of numbers

Reply on news.ycombinator.com