diff --git a/.gitattributes b/.gitattributes index 2c596aa58456a4c41df348f8f65bc6f8a5c174c6..24f96a28797e3079cb5ba6bbb705206cbd083d7e 100644 --- a/.gitattributes +++ b/.gitattributes @@ -241,3 +241,27 @@ embeddings/WeFighting.preview.png filter=lfs diff=lfs merge=lfs -text embeddings/WebbedX.preview.png filter=lfs diff=lfs merge=lfs -text embeddings/WrenchWench.preview.png filter=lfs diff=lfs merge=lfs -text embeddings/ZSA_IL_Pos.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/BS_BoobiesSDXL.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/CaughtShowering.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/FAE_AssOnGlass.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/FAE_BreastsOnGlass.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/FAE_LeaningOnViewer.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/GetOut.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/HDA_AmazonPositionXL_v1.2.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/HDA_Artificial_Vagina.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/HDA_Ellen[[:space:]]Joe.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/HDA_Prostration.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/HDA_Wind_lift.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/ILXLneg.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/ILXLpos.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/LiadoraDescILL.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/Panacea.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/ShowOffFinger.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/SmthILLS.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/TSTST0.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/Tr1sh.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/illustrious_renaissance_negative_v2.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/m4z0h-IL.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/neghand.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/solo-femboy-IL0.1.preview.png filter=lfs diff=lfs merge=lfs -text +embeddings/wendyGNRCemb.preview.png filter=lfs diff=lfs merge=lfs -text diff --git a/embeddings/BS_BoobiesSDXL.html b/embeddings/BS_BoobiesSDXL.html new file mode 100644 index 0000000000000000000000000000000000000000..3fb6129e74ee65f823e51fa61549bc33f071f535 --- /dev/null +++ b/embeddings/BS_BoobiesSDXL.html @@ -0,0 +1,79 @@ + +
+ + + + +
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Well I decided to create an embedding for a SDXL/Pony/Illustrious just based to a prompt, there is no training at all, just a simple prompt detailing woman's breast. Done with: https://github.com/klimaleksus/stable-diffusion-webui-embedding-merge
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Chosen character is caught in the shower.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
This embedding generates a pose where character is pressing her ass against glass.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
This embedding generates a pose where character is pressing her breasts against glass.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
This embedding generates a pose where character is pressing against the viewer.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
NSFW embedding that puts a chosen character (female only at the moment) in a bed room naked yelling at someone to get out.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Remember to put the .pt / .safetensors file in the "embeddings" folder
there is no need to add 1boy to positive prompts
If you appreciate my work and want to ✨SUPPORT IT✨, you can buy me a coffee ☕ on BOOTSY or UNIFANS
⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️
Check out all HDA embeds
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Updated version for Illu and accessible to all users
Remember to put the .pt / .safetensors file in the "embeddings" folder
⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️
Check out all HDA embeds
and
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️
Check out all HDA embeds
and if you want to discover all the previews

Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Use HDA_Prostration in your positive prompt to generate images.
Remember to put the .pt / .safetensors file in the "embeddings" folder
If you appreciate my work and want to ✨SUPPORT IT✨, you can buy me a coffee ☕Unifans
and if you want to discover all the previews

Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
✨SUPPORT ME✨ to get access to over 200 supporter-only embeds
⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️
Check out all HDA embeds
and if you want to discover all the previews
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Negative Illustrious Embedding is a carefully selected negative embedding tailored for the Illustrious model family. It helps eliminate common rendering issues such as distorted anatomy, blurry features, washed-out colors, and messy compositions.
In my example, I used the ✨JANKU v3.0 - NoobAI EPS + RouWei NSFW Illustrious XL✨ model. This model includes its own embeddings, which work really well in practice. I was simply curious to see how the model would respond under these conditions and how effectively it would apply those embeddings.
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Positive Illustrious Embedding is an embedding specifically designed for the Illustrious family of models that enhances image clarity, expression, and artistic accuracy, delivering glossy smoothness and sharper facial features with a rich, vibrant color palette suitable for portraits and anime illustrations.
In my example, I used the ✨JANKU v3.0 - NoobAI EPS + RouWei NSFW Illustrious XL✨ model. This model includes its own embeddings, which work really well in practice. I was simply curious to see how the model would respond under these conditions and how effectively it would apply those embeddings.

Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Warning: Those don't work with Civitai On-site generator. Once again, i don't understand why some embeddings are ok and some others are not. It feels like it due to their size: if they go over a limit of number of tokens, they are just ignored 😥
I still haven't done the LoRA for my OC, but nonetheless, here are the embeddings, i have used until now to generate her. They work nicely, the only drawbacks is that they are a LOT of tokens internally: a LoRA would allow having a prompt more focused on everything else in the picture composition.
LiadoraDescILL (39 tokens): The base description of Liadora (hair and eyes colors, the tattoo and horns)
LiadoraOutfitILL (54 tokens): Her standard outfit
LiadoraNegILL (26 tokens): The "fix" negative to avoid some influence from 2B (white hair and Yorha outfit)
(I have added the number of tokens corresponding to each embeddings. Remember that for any SDXL models and derivative, a prompt is split and ingested 75 by 75 tokens).
A base prompt:
Positive: 1girl, LiadoraDescILL, LiadoraOutfitILLNegative: IllusN3g, LiadoraNegILLThe result (bonus trick: if needed, you can increase the weight of the embedding, like LiadoraNegILL:1.1)

A more complex prompt may need to split the embeddings into two (at least) chunks manually using BREAK to avoid side effects (and that's why i'll need to make a LoRA out of those 😅):
1girl, solo, LiadoraDescILL, sitting on a throne, indoor, castle, (conjuring a blazing purple fireball with one hand)
+BREAK
+LiadoraOutfitILL, from below
+BREAK
+volumetric lighting, black background, simple background, vibrant, stunningly beautiful, crisp, detailed, sleek, ultramodern, suble shadows, high contrast, cinematic, ultra detailed, intricate, professional

Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Amy Dallon in her Panacea Robes
Just the Character in costume.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Female character leans forward and shows off her cleavage while flipping the bird.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
This CAN enhance just about ANY ANIME model to do some really cool stuff.
the NAI one was rolled off NoobHoshi or something -- i can't remember what it was called, but you'll see what it is when it comes up.
The two main Illustrious ones are built for my upcoming model. Replicant.
That does NOT mean it won't work on any Illustrious, but be careful: TEXT ENCODERS VARY.
Forge/A111 sometimes I can use PDXL and ILLS ones if it's consistent information like my characters, but that's my laziness. MOST OF THE TIME STICK TO THE BASE MODEL (tho don't use ILLS/PDXL on XL it may crack and burn).
Discord: https://discord.gg/HhBSvM9gBY
Twitch: https://twitch.tv/duskfallcrew
Arc En Ciel: https://arcenciel.io/users/77
Bluesky: https://bsky.app/profile/duskfallcrew.bsky.social
Sponsor models here: https://ko-fi.com/duskfallcrew/commissions
Find backups here: https://huggingface.co/EarthnDusk/
(CopyleftrightlefttheairforcerepthemilitarystratLETSPLAYDDR (All text & Models on this profile are produced by - 2025, 0FTH3N1GHT Productions, Ktiseos Nyx)
Test it out on Tensor: https://tensor.art/u/611011406535381539
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Here comes a "new" stupid idea of mine that will probably never be useful to anyone but me 😅
Warning beforehand: Those embeddings have been created using an extension in A1111 and tested on the same platform. I haven't tested them with ComfyUI but i have tested those with Civitai on-site generator... and they WILL not work. That's why i have disabled on-site generation with them.
TSTST are a set of embeddings containing a full prompt (~75 tokens) to generate a test picture with a single embedding word. I'll upload more as I build them, but for now, here is the first one. They are built with Illustrious in mind, but they will do something with any SDXL model of course. I am releasing those here since I'll be using them in uploads.
TSTST0: This one should result in a close up picture of a slightly cyberpunk girl partially out of frame with rainbow eyes. Anything else and something went wrong. 😱
TSTST1: A gothic elf in lingerie under the rain with a black background. Heavy on the purple grading. If some gold/orange comes out, included "golden hour glow" messed up.
TSTST2: The Angel 😉 (well, with a somewhat less detailled prompt, 75 tokens limit quicked in)
TSTST3: A bishounen laying upside-down on a bed. This one IS a real hard test => base Yolo coming with Adetailer WILL fail to detect the face. Switch to a V11 and/or lower detection threshold to get it right 🤞
TSTST4: A generic landscape but a few elements can be tricky: a dragon flying may or may not appear, reflection in the water can be false and if either the wooden cabin or the wheat field is missing, something went wrong.
PS: yes, i'll probably convert most of my usual "showcase" pictures into those, but i'll stop at 5 for now 😄
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Trish, my first original character, even tough she's not very original.
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Illustrious Renaissance v2.0 (unreleased - under testing phase)
Toxic Muse Negative embedding v1.0 is a handcrafted negative embedding designed specifically for my latest generation of models.
Unlike general-purpose embeds, this one is tailored to enhance prompt control, reduce noise, and preserve artistic consistency across highly stylized, detailed compositions.
✅ Reduces anatomical distortions (fingers, eyes, limbs, symmetry)
✅ Removes unwanted artifacts (blur, watermark traces, color mush, edge noise)
✅ Improves prompt adherence and style purity
✅ Enhances both NSFW and SFW scenes, especially in multi-subject or complex settings
Just add the embedding to your Negative Prompt. No trigger word is required.
This embedding is model-specific and fully tuned to Toxic Muse v2.0 & Illustrious Renaissance v2.
Probably, but I haven't tested it with other models
Without Toxic Muse Negative embedding v1.0
-------------------------------------------------------------------------------------------------------------
With Toxic Muse Negative embedding v1.0

These samples were generated using the same prompt and seed — the only difference is the presence of the embedding in the negative prompt.
Notice the boost in clarity, edge definition, and style consistency.
If you're using my v2 models, this is the official negative embedding designed to help them shine.
More stable, more beautiful, and with fewer headaches.

Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
This embedding was designed to produce a blend of different expressions, aiming for a conflicting expression of smug, shame, desire and fear. That fear you have of your own desire when you know you are about to get addicted to some: thing, taste, or act. But you are also ashamed of it, yet maybe also smug that you can finally indulge your desire.
In general there is no need for another triggerword. For better results, I find (m4z0h:1.5) works wonders.
However, if you want to highlight expressions, aspects or concepts: heart-symbol pupils, saliva, drooling, tongue, smug, desire, ashamed, awe etc... it should respond accordingly.
While this was designed to tame a specific tag while introducing conflicting expression concepts to allow it to be more flexible with expressions. That taming doesn't always work and it can easily go too far into nsfw, so use "rating_safe" if you are after safe images.
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Negative Illustrious Embedding is a carefully selected negative embedding tailored for the Illustrious model family. It helps eliminate common rendering issues such as distorted anatomy, blurry features, washed-out colors, and messy compositions.

Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
an embedding to generate anime femboys/traps in illustrious so newbies who can't into japanese can gen them
put it in your embeddings folder. refresh the embeddings in the webui. prompt "solo-femboy-IL0.1, general, outdoors" or whatever
bro you have to write solo-femboy-IL0.1. it's not "femboy". it's solo-femboy-IL0.1. it's the name of the embedding file you download
the embedding is literally "1boy, otoko no ko, solo, male focus" created from illustrious v0.1. weights are more or less the same in other models so it's not exact but it should work
+
Use the model without crediting the creator
Sell images they generate
Run on services that generate images for money
Run on Civitai
Share merges using this model
Sell this model or merges using this model
Have different permissions when sharing merges
Wendy is a fake person. She has been created using no images of real people, and can be used freely for any SFW and NSFW applications. Enjoy!Wendy is the main reason why i made the SD1.5 to SDXL embedding converter 😍
All the nice embeddings by @Manuel_Kidney are very cool but i wanted to try some of them on SDXL/Illustrious XL.
Sadly, those models are more biased than SD1.5 and depending on the choosen model, the results vary A LOT with only an embedding: as a proof, the base converted embedding sometimes don't provide a dark haired beauty if the model is biased toward blondes...
That's why this embedding has been appended with ", dark hair" using the Embedding Merge extension after conversion to help. But it was not enough, so, there is also a LoRA!
Please note that Illustrious models tend to generate young characters, don't forget to add a negative prompt toward young character if needed.
+