Join our Discord to see more images and upcoming updates!
Sexy Mouth & Cumshot v2 XL
This LoRA can improve the mouth generation by adding more emotions. Although BigLust can generate such content it was dull and felt generic. With this LoRA the mouth looks more sensual and it is easier to control. The training set is increased to 300 images.
Model 1: Big Lust v1.0 (SDXL) Brighter images (I have used this for training)
Model 2: Big Lust v1.5 (SDXL) Darker images with minor quality loss
SDXL ≠ Pony You should use SDXL models for generation if you want realistic results.
Training Resolution
Resolution 1: 832x1248, 2:3, Portrait
Resolution 2: 1024x1024, 1:1, Square
Resolution 3: 1248x832, 1:1, Landscape
Training time: 11.5h (RTX 4090)
For tags see v1 or the LoRA metdata.
Sexy Mouth & Cumshot v1
Training Data
This LoRA was trained with images from face and upper body with open and closed mouth, tongue out and cumshots. All images are manually edited, enhanced and tagged to get better results.
Model: Lazymix Real Amateur 3.0a
Note: For testing purpose I have also trained the LoRA on Lazymix v4.0 but found the results with v3.0a slightly better. You can generate with v4.0 but since the merge is different compared to v3.0a you will get different results.
Image count: 230
Resolution 1: 627x941, 2:3, Portrait, 41%
Resolution 2: 768x768, 1:1, Square, 40%
Resolution 3: 941x627, 1:1, Landscape, 19%
Tags
Keep in mind: Tags raise the chance to get what you want. But depending on the amount of training images containing this tags some tags might need more luck than others to work.
Only use a single main tag or you might mix things up.
Main tag
sexymouth
Mouth & Eyes
close up, extreme close up - Zoom level of the face (add to negative for body pose)
wide open mouth, open mouth, parted lips, closed mouth - self explanatory
tongue out - Sticking tongue out
open eyes, closed eyes - self explanatory
Objects & face effects
licking [object] - Currently only trained with following objects: cock, cocks
cumshot - Cum on face
spurt - Live cumshot, not realy working so far (need more images)
saliva - Wet mouth with saliva
makeup - Woman was wearing makeup, Seems to generate colored eye shadows
Other tags
front view - Best with portrait resolution
side view - Best with landscape resolution
view from above - Kneeling on floor, waiting for your cum
blurry background - Used quite often for unknown backgrounds
Example prompts
sexymouth, open mouth, tongue out, photo, woman, brown hair, medium breast, open eyes, makeup, front view, cumshot, gray wall in background
sexymouth, wide open mouth, tongue out, close up, photo, woman, blond hair, open eyes, dildo, makeup, front view, saliva, wearing outfit, indoor, painting on wall
sexymouth, open mouth, tongue out, close up, photo, woman, blond hair, open eyes, licking cock, makeup, front view, wearing orange dress, blurry background
sexymouth, parted lips, close up, photo, woman, blond hair, open eyes, front view, cumshot, ear rings
sexymouth, open mouth, tongue out, close up, photo, woman, dark hair, open eyes, side view, saliva, necklace, hair clip, blurry background
sexymouth, wide open mouth, close up, photo, woman, blond hair, small breast, closed eyes, cock, makeup, side view, cumshot, spurt, necklace, empty room, white wall, wooden floorDescription
first release
FAQ
Comments (12)
Are these local output, or a service like Mage.space?
All images are generated local with Automatic1111
@Lucifie More advanced than what I'm getting. I see "backbone_factor" in the code, are you running FreeU?
@camerique701 I think it was enabled on a few images.
samples look excellent, thanks!
This looks based on XL, not SD 1.5- the file size is bigger than 144 mb
It's 1.5 with higher training settings to get better details so the file is bigger.
@Lucifie can’t you do a 144mb version?
@portraitartist999132 In Kohya you can reduce the size if you want even down to 8 mb. I have tried this before but the smaller the file the more details are lost and you need way more images to get a good one. I prefer quality over file size.
If this is for XL, it should be marked in the title as an XL LoRA. This is good, but I need a version for SD 1.5. Can you make one please?
See my other comment.
amazing! can you do this one for pony?










