Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

All of it is directly applicable to LLMs.

Textual Inversion/Soft Prompts are not made "obsolete" at all by Loras. Textual Inversion is far faster to train, extremely effective on newer models like SDXL which have better clip encoding, and has neat properties of not messing with the general weights of the rest of the model, which is sometimes desirable.



Consider applying for YC's Summer 2026 batch! Applications are open till May 4

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

Search: