Illustrious conversion + Pony remake
Haven't revisited this LoRA in a long time, the original version was very rigid and didn't blend well with other LoRAs. With this update it should be a lot more customisable with the pose.
This is a pov sex position Lora to have the character bound to a chair.
Core prompts
Pov Bondage Chair default,
Pov Bondage Chair legs up,
Pov Bondage Chair legs down,
Customisation prompts
1girl, 1boy, pov,
1girl, solo,
restrained,
arms up,
arms by side,
from behind, butt,
from above,
dutch angle,
imminent sex,
sex,
imminent anal sex,
anal sex,
internal view, x-ray,
cumming, creampie, ejaculation, trembling,
pussy,
nipples,
penis,
anus,
fully clothed,
open clothes,
naked,
breast grab,
thigh grab,
butt grab,
Description
This is a pov sex position Lora to have the character bound to a chair
Prompts
Pov Bondage Chair,
no sex, (add breasts exposed, lower half exposed or fully exposed prompts to customise)
pressed against vagina,
sex,
deep sex,
pressed against anus,
anal sex,
deep anal sex,
Creampie,
FAQ
Comments (8)
Thank you for your nice LoRA!
I'd like to use your dataset for training SD1.5, what do you think of it?
Yeah sure thing go for it, thanks for asking beforehand! The only thing I would ask is to either link to my LoRA in the description or put mine as a suggested resource. Or just link to my profile whichever is easiest
@Oscarc Thanh you for replying! I understand your requirements about link and resource. After some test generations, I would publish the LoRA.
@unexpectedlyprovided Hello. I never trained a LORA but I'm just curious about. I see the dataset is done with AI images. Made how?
@settima_ai hey there, I don't think I make LoRAs the same way most people do so there is probably much better ways to get training data out there, as I've been winging it this whole time. I generate pictures on the civitai website using different prompts until I get enough pictures for the pose I'm currently trying to make a LoRA for (usually takes quite a bit of trial and error as some poses are hard to do in AI). Then I train the LoRA (also through the website) with the data set I generated, making sure everything that needs to be prompted is added to the pictures after the auto prompt feature runs. Then if everything went right with the training the LoRA should function as a short hand, making it a lot easier to get the AI to generate pictures in a specific pose. Hope this helps!
@unexpectedlyprovided did you make one?
@davelyin47536 here, sorry for it's the old version.
https://civitai.com/models/612894?modelVersionId=685145
@davelyin47536 here, sorry for the old version (not remake ver.)
https://civitai.com/models/612894?modelVersionId=685145
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.



















