The strength of Lora is recommended to be around 0.5
I made it by request, but the training image is biased, so please keep the intensity low.
Description
FAQ
Comments (6)
WHYYYYY do you NEED this ?
can't you prompt it ?
I made this because there was a request.
@bolero537 well i can't argue then... XD
id like to imagine getting paid for that.
now that would be a scam talent !
@Le_Fourbe Using something like this also helps with consistency.
@MidSpin while this is true, it's much beter to learn the basics before downloading useless bloat.
medium hair + side ponytail have the EXACT same effect. (Tested on KonpaEVO mix and NTR mix with quality/negative tags)
photobashing the ponytail in img2img is also consistent. controlnet reference IP adapter is also a good trick to get the same face.
if you look for a style, use either mix of style lora and art related prompt, this is much more useful at large. you don't want to have a lora for every words do you ?
@Le_Fourbe While learning the basics can bring you everything you need, (As in all things in life) Rome wasn't built in a day. Just yesterday, I was having a problem with getting a consistent highly complex hairstyle to stay in style for a series of pictures. While I used inpaint, regenning, and seed manipulation to make it work, If someone made a Lora for the hair, It would've cut a 3 hours job to an hour long. But again, I feel we are in agreement on most of this already.
