This is model uses the exact same input as bimbostyleFour and bimbostyleFourU.
I was trained on SD_v1.5 which hopefully means that it's better applicable to other .ckpt in comparison to the normal "bimbostyleFour" and will trsy to emulate the style of the ckpt more than the deliberateV2 one
This lora is very responsive to body size prompts i.e. large breasts/wide hips/thick lips
And can also be used to bimbofie an existing model or lora
Combining this Lora together with One/Two/Three/Four in a "draw style" ckpt give great results and i can only recommend trying it out. (i don't know if the same applies to "real" ones as well)
can be mixed with every other bimbostyle
If you mix my models, i would recommend to always put the stronger weighted Lora in the front. i.e.
<lora:bimbostyleThree:0.5>:<lora:bimbostyleTwo:0.3>
Also i would go above a combined strength of 1 and from experience i would say .75-.9 is the sweet spot total strength
Description
FAQ
Comments (11)
I really like this. Did you do something different between this kind of process VS BimbostyleTwo? I like the thinner smaller but still bolt on breasts from Bombostyletwo but the results are not successful as this Lora. Basically Pornstar style like Nicollete Shea, or Lmonies
I made around 100 more loras between my bimbostyleTwo and this one, so my training dataset and tagging is much better this time around :D,
I'll can try and make a bimbostyle for smaller bolted-on breasts, but for now i'll t ry and update all the later one to the SD_v1.5
when will you make a photo based model?
not quite sure what you mean by that, all my lora are trained on real photos.
I'll upload my person specific Loras as soon as i'm sure that my lawmaker won't fuck me over.
@caphalorthrow Whos causing trouble for AI?
@findingavarice521 the European union with their proposed "The Artificial Intelligence (AI) Act" as soon as that is passed i'll now where they stand. Also the current laws in my country doe not quite secure me from Repercussion if anyone wants to try.
@caphalorthrow I meant that the model you trained it on. I haven't trained LORAs, but I mean that the checkpoint is not revanimated or deliberate, but something like URPM
@faketitslovr ahh -.-" my bad.
I retrained them all now on SD_v1..5, but i can also try an maybe get it on URPM as well.
@faketitslovr
I added styleThree and Five trained on URPM on my hugginface if you want to give them a spin
https://huggingface.co/caphalorthrow/Bimbos/blob/main/bimbostyleFiveURPM.safetensors
https://huggingface.co/caphalorthrow/Bimbos/blob/main/bimbostyleThreeURPM.safetensors
@caphalorthrow sweet. I loved five!
@faketitslovr
glad you like it, sadly my body slicing didn't yield to great results so the asia bimbostyle is still a bit away