See also the LoRa
====
A little textual inversion trained on pictures of Halle Berry wearing a fancy bikini and huge sunglasses.
I retrained a whole embedding with the same pictures, but the result was efficient enough. I then decided to merge it with the first version through the Embediding Inspector extension with ("v/2" as eval expression).
Go ahead and upload yours!
Your query returned no results β please try removing some filters or trying a different term.