Hacker News new | past | comments | ask | show | jobs | submit login

It seems like it could be any transformer, which is exciting now that even in imaging gradient transformers are all the rage. But ideally we'd need to see this result in other transformers (but I have a hard time seeing why it wouldn't be the case).



At the very least it could be used to reduce the requirements and speed up the prompt recognition step(s) of image-based generative AI.

"Stable Diffusion 3 XS" will use ternary? Here's to hoping :)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: