A Flux LoRA that will make any women in the image more busty.
Important: This model is 100% fictitious and is not based on images of any real person. All dataset images originated with images generated from still image checkpoints and that evolved from there using only AI rendering. No images (neither actual nor AI-generated) of any real person were ever involved in the making of this model.
PLEASE READ the "about this version" statement on the right side of the screen. There are lots of details about the latest version including prompt tips and information.
The model was trained on clothed women only, so this model won't help with Pony-style nude images. (includes nude/topless since version 2.0) It works well with all kinds of outfits from loose and baggy shirts to skimpy bikinis. There is no trigger word needed. Just add the LoRA to your pipeline and... instant cup sizes.
The results are mostly in the "large" or "very large" range with occasional results below or above that. This is relative to people in real life, not necessarily to other AI images you may find here on Civitai. The look is usually natural but not always.
How to use:
Add the LoRA to your image creation
There is no step 2.
Possible issues:
Sometimes the women have unusually hard nipples poking through the outfit. I haven't found any way to tweak my prompts to remove that, so if you see it you're stuck with using a different seed or else scale back the LoRA intensity to 0.5 or 0.6.
While no trigger word is needed, you can ask for even bigger by specifying "busty", "large breasts", "huge breasts", or "oversized breasts" for more extreme results.
For reference, all of the sample images were created with Flux Dev, fp8, on my personal machine using comfyui. Each image has the comfyui workflow embedded, so if you want to re-create it you can download the full size image and import directly into your own comfyui. The workflow I like to use for general purpose Flux generation is the one I published here: https://civarchive.com/models/906003/flux-easy-multi-controlnet-selector-workflow-for-comfyui
That workflow has a plain Flux processing option so you don't have to bother with controlnets if you don't want to.
Enjoy, and please don't forget to post any images you make with this LoRA back to this page. I'd love to see what you can create!
Description
FAQ
Comments (6)
Great job! For some reason this didn't appear in Civitai's front page when I searched for flux loras of the past week.
Anyway, I use Forge, not Comfy. Have you tried it there? My results with it are not great. I'm not sure if my prompt/settings are bad or if Forge is the problem. Can you give an example prompt with your settings?
I've never tried Forge so I don't have any direct experience that can help. Here is the info for a recent photo from the series I posted with women getting out of sports cars...
Prompt:
A gorgeous fashion model blonde woman with long sexy legs is sitting in a fancy sports car. The door is open and we can see her beautiful legs. She is wearing silk stockings with a very short tight dress high heel shoes. Perfect muscular legs. Close-up legs. Extreme low angle shot. Glamorous lipstick and makeup.
Steps: 20
Size: 1024x800
Seed: 138890417184552
Flux guidance: 3.5
LoRA: BustyWomen Flux at 1.0 strength
I also used another LoRA to get a more athletic look, set to 0.4 strength: https://civitai.com/models/664534/female-body-builder-muscle-physique-flux (I did not use any keywords in my prompt as you can see)
Sampler: Euler
Scheduler: Beta
I think those are all the things I can configure in ComfyUI. Those parameters gave me this image: https://civitai.com/images/39558728
Good luck! I hope you can make this model work, I've been having a lot of fun with it.
@darkroast175696 Yes, this helps. My number of steps was wrong and my guidance was too low. I tried it with the STOIQO checkpoint, and this lora doesn't work very well with it.
@Solus I generally just use vanilla Flux Dev 1 with the fp8 models because I just have an Nvidia 3060 and it can't handle anything bigger than that.
@Solus I did a local install of Forge and got flux working with the flux1-dev-bnb-nf4 checkpoint since it builds in the vae and text encoders for less VRAM usage. I made a few images with this LoRA and they seemed to work pretty well. I also see you posted a few yourself so it looks like you got things working. I hope you enjoy working with it.
@darkroast175696 Yes, I got it working. Once you posted those settings I was able to figure out what I did incorrectly. I also only use the fp8 version. My GPU could theoretically handle fp16, but it cries and whines when I try!