単眼娘が出せるLora(Loha)です

プロンプトには tangan, cyclopsを追加してください

学習に使用したモデルはanything4.5です

Lora (Loha) who can produce cyclops girl(tangan musume).

Please add tangan, cyclops to the prompt

The model used for training is anything4.5

This model can be used to create icon light and silhouette effects, suitable for most large models. At present, it is the first version, and there are still many shortcomings. I am looking forward to your feedback during the use process.

这个模型版本主要有助于增强光效和剪影效果，在多数大模型上都有效，并且产生不同的效果。建议把两个触发词一起加上，并加上自己想要的背景，酷炫的效果就来了。

测试人物类效果比较显著；测试物体类，目前是剑的效果不理想，无法得到单独的剑的剪影，始终有人的剪影出现。

欢迎大家使用，希望得到大家使用过程中的反馈，以期做出更好的版本。

Get early access to build and test build, be able to try all epochs and test them by yourself on Patreon or contact me for support on Discord

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)

• CFG Scale : 5 to 10

• Lycoris weight for txt2img: reccomended 0.35

Those are a Lycoris, to use it you need to install the proper extension on Stable Diffusion.

My advice is to start from the prompt of the posted images on civitai, or in any case use most of the keywords with 1.2-1.3 weight.

In general if you want a classic ironman suit you just need to use the keyword "wearing irongirl_suit" maybe with a weight if necessary.

If you want a mix like irongirl|ballet or irongirl|spider you should use both keywords, in this cases i have an important advice, play with weights! mostly when adding a particular color to the suit. I make you an example, most of the training ballet dress images are white, most of the training irongirl suit are red, so if you write "(wearing ballet dress with tutu, wearing white irongirl_suit:1.3)" SD will probably generate a normal white ballet dress, so in this case it's better to lower the weight for the ballet dress to have the mix you want, like "(wearing ballet dress with tutu:1.1), (wearing white irongirl_suit:1.3)". Opposite case if you want a red irongirl ballet dress suit.

Enjoy!

If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: https://www.patreon.com/betterCuteAsians

Get early access to build and test build, be able to try all epochs and test them by yourself on Patreon or contact me for support on Discord

My advice is to start with posted images prompt.

Also, for each model, look on version description and on keywords on the right panel.

Enjoy!

If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: https://www.patreon.com/betterCuteAsians

英语(English): This model does not come with clothing information, but it has correctly memorized the body data through regularization, so the character proportions are perfect. You can dress her in any clothes. A weight of around 0.6 is recommended.

中文(Chinese): 本模型不带有服装信息，但是通过正则化正确地记忆了身体数据，因此人物比例完美。你可以给她穿上任何衣服。推荐权重0.6左右。

日语(Japanese): このモデルは服の情報を持っていませんが、正則化を通じて体のデータを正しく記憶しているため、キャラクターの比例は完璧です。彼女にはどんな服でも着せることができます。推奨される重みは約0.6です。

韩语(Korean): 이 모델은 의류 정보를 포함하고 있지 않지만, 정규화를 통해 몸 데이터를 올바르게 기억하고 있으므로 캐릭터의 비율은 완벽합니다. 그녀에게 어떤 옷이든 입힐 수 있습니다. 권장 가중치는 약 0.6입니다.

There's a few rainbow/colorful models, but I wanted one that could replicate the rainbow drip/splatter effect and rainbow chalk effect, and also work with realistic checkpoints. Give it a try. Add the rainbowpunk keyword to activate it.

I intended it to be fairly general, but might be useful for making Pride images too.

Because it is based on LoRA but is trained with Conv2D
Conv2D requires need extensions
https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris
or
https://github.com/kohya-ss/sd-webui-additional-networks

训练测试，质量比较一般，推荐权重0.5-0.7，基模Anything5，有时候上衣服会变成别的颜色，有的图是用ControlNet-Tile以及Ultimate SD Upscale脚本在图生图里面放大过。

Training test, the quality is not very will, recommended weight is 0.5-0.7, the base model is Anything5, sometimes the clothes will change to other colors, some images is enlarged in the Image2Image with ControlNet-Tile and the Ultimate SD Upscale script.

英语(English): This model does not come with clothing information, but it has correctly memorized the body data through regularization, so the character proportions are perfect. You can dress her in any clothes. A weight of around 0.6 is recommended.

中文(Chinese): 本模型不带有服装信息，但是通过正则化正确地记忆了身体数据，因此人物比例完美。你可以给她穿上任何衣服。推荐权重0.6左右。

日语(Japanese): このモデルは服の情報を持っていませんが、正則化を通じて体のデータを正しく記憶しているため、キャラクターの比例は完璧です。彼女にはどんな服でも着せることができます。推奨される重みは約0.6です。

韩语(Korean): 이 모델은 의류 정보를 포함하고 있지 않지만, 정규화를 통해 몸 데이터를 올바르게 기억하고 있으므로 캐릭터의 비율은 완벽합니다. 그녀에게 어떤 옷이든 입힐 수 있습니다. 권장 가중치는 약 0.6입니다.

英语(English): This model does not come with clothing information, but it has correctly memorized the body data through regularization, so the character proportions are perfect. You can dress her in any clothes. A weight of around 0.6 is recommended.

中文(Chinese): 本模型不带有服装信息，但是通过正则化正确地记忆了身体数据，因此人物比例完美。你可以给她穿上任何衣服。推荐权重0.6左右。

日语(Japanese): このモデルは服の情報を持っていませんが、正則化を通じて体のデータを正しく記憶しているため、キャラクターの比例は完璧です。彼女にはどんな服でも着せることができます。推奨される重みは約0.6です。

韩语(Korean): 이 모델은 의류 정보를 포함하고 있지 않지만, 정규화를 통해 몸 데이터를 올바르게 기억하고 있으므로 캐릭터의 비율은 완벽합니다. 그녀에게 어떤 옷이든 입힐 수 있습니다. 권장 가중치는 약 0.6입니다.

Be not afraid!
-A creature you should certainly be afraid of.

This LyCoris produces many winged humanoid angels or crazy eyeball ring monsters depending on the prompts used.

"too many eyes, no humans" for eyeball type angels with the rings and such

"wings" or "multiple wings" for your more humanoid angels. Adding "no humans" tends to cover them in feathers or fur.

👑 Alissa White-Gluz Lycoris 👑 V5 - Realistic

🔧Model base: AngrA_RealFlex_v02

_________

📝🌟 Trigger: AlissaWhiteGluzQuiron woman, blue hair, <lyco:quiron_AlissaWhiteGluz_Lycoris_v5:1.0>

_________

📝 Variation: <lyco:quiron_AlissaWhiteGluz_Lycoris_v5:Y>, Y=0.6 to 1.0

📝 Denoising strength: 0.30 to 0.65

_________

🗒️💡Sample:

(best quality:1.1), (masterpiece:1.2), (realistic:1.2), realistic lighting, trending on Artstation, photoshoot of a AlissaWhiteGluztQuiron woman, blue hair, <lyco:quiron_AlissaWhiteGluz_Lycoris_v5:1.0>,

_________

analog style, modelshoot style, portrait of AlissaWhiteGluzQuiron woman, wearing a biomechanical organism on a rampage, bold, bright colours, <lyco:quiron_AlissaWhiteGluz_Lycoris_v5:1.0>,

_________

🚀 This LORA likes: (best quality:1.1), (masterpiece:1.2), (realistic:1.2), (detailed:1.1),

👇 Negative Prompt: paintings, sketches, (worst quality:1.2), (low quality:1.2), (normal quality:1.2), lowres,

💡Sample steps: 18 to 35

💡CFG Scale: 5 to 8.5

Hire. fix: ok

👇Restore faces: Yes /No

____________

☕ If you feel like supporting my work, please consider visiting my Beer / BuyMeaCoffe ( https://www.buymeacoffee.com/quiron ) ☕

🌟☕If you like my work please support ( https://www.patreon.com/VisionsofAi )☕🌟

________________________

______ Nerd Info: ______

image_count: 4420

num_repeats: 1

shuffle_caption: False

keep_tokens: 0

caption_dropout_rate: 0.0

caption_dropout_every_n_epoches: 0

caption_tag_dropout_rate: 0.0

color_aug: False

flip_aug: False

face_crop_aug_range: None

random_crop: False

token_warmup_min: 1,

token_warmup_step: 0,

is_reg: True

class_tokens: woman

caption_extension: .caption

import network module: lycoris.kohya

Using rank adaptation algo: lora

Apply different lora dim for conv layer

Conv Dim: 1, Linear Dim: 128

Use Dropout value: 0.0

Create LyCORIS Module

create LyCORIS for Text Encoder: 72 modules.

Create LyCORIS Module

create LyCORIS for U-Net: 278 modules.

enable LyCORIS for text encoder

enable LyCORIS for U-Net

prepare optimizer, data loader etc.

num reg images: 4420

num batches oer epoch / 1epoch: 666000

LoHA trained on Coven & IG Leblanc from League of Legends

IG requested by @1797437344610

Coven requested by @Cinders

models used

anime: YabaL_Mix V3, MeinaAlter

2,5D: HELLmix(Picture3,10,11), DarkSun(picture 2), CamelliaMIx_2.5D_V2(picture 4,5,6,8,9), MeinaUnreal

realistic: fantasticmix(picture 7)

Coven Leblanc: coven leblanc, 1girl,

optional tags: black dress, feather trim, fur trim, long hair,

IG Leblanc: ig leblanc, 1girl,

optional: short dress, cape, elbow gloves, hair ornament, white dress, short hair,

negative embeddings: embedding:bad-hands-5, embedding:ng_deepnegative_v1_75t, embedding:bad-artist, embedding:EasyNegative

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• VAE: vae-ft-mse-840000-ema-pruned (for realistic models & 2,5D), kl-f8-anime2, blessed2.vae (for anime models)

• Sampler: DPM++ SDE Karras or DPM++ 2M Karras

• Steps: 20 to 35

• Clipskip: 2

• Upscaler : img2img in comfyui using 4x-UltraMix_Smooth

• CFG Scale : 5 to 9 (mostly 7 and 8)

• LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.8)

• Denoise strength for img2img: 0.4 to 0.7

for the pictures i used comfyui (thats why there are no

Please check out my Ko-Fi page if you are interested in supporting me.

LoRA commissions open! -> Fiverr

This is a LoHa LoRA! You need the extension! Why? They do a lot better with multi-concepts.

This is retrained from PointOfKnowReturn dataset! He was kind enough to share his dataset.

Trigger words for V1.20:

positive: <lora:mushroomPeoplev120:0.75>(mush_woman, blue mushroom)

You can also try to prompt for size like "big mushroom", "medium mushroom", "small mushroom", you can prompt for colors if you want and it does colors relatively well and also "glowing".

Lower Lora weight to generate men.

Follow the steps in the Model Version notes to reproduce my images.

Sample images generated with Nothing, DeimosAnime and .

All negative embeds should be easy to find. and Prompts should be included in the sample images! Click the little icon in the bottom right to see it.

It was trained on 62 512x512 images provided by thesauceer, 20 epochs using Animefull-final-pruned for the Training Model.

Danbooru tags generated by Waifu Diffusion 1.4 Tagger.

Kagura (fully grown) from Senran Kagura

• Trained with anime (full-final-pruned) model.

• Recommended LYCORIS setting; dyn=8:te=0.5:unet=0.5

• 2 versions; 6 and 10 epochs.

Please check out my Ko-Fi page if you are interested in supporting me.

LoRA commissions open! -> Fiverr

This is a LoHa LoRA! You need the extension! Why? They do a lot better with multi-concepts.

I've attempted this multiple times, this is the best I could come with.

Trigger words for V2.20 - You should probably avoid gens with the character too far away!:

You can try "sleevemesh" for armwear, "pantynet" for pantyhose, "sheertop" for a top (probably needs to add "nipples" to it) , "nettedthighs" for thighhighs.

NOTE: From my experience testing it, looks like thighhighs and top needs higher weights.

Follow the steps in the Model Version notes to reproduce my images.

Sample images generated with Hassaku.

All negative embeds should be easy to find. and Prompts should be included in the sample images! Click the little icon in the bottom right to see it.

It was trained on 120 512x512 images borrowed from Unstable Diffusion discord, 20 epochs using Animefull-final-pruned for the Training Model.

Danbooru tags generated by Waifu Diffusion 1.4 Tagger and fine tuned manually.

Because no one its creating the young version I thought it was a good shot to do it, so...

As a note all my Photorealistic characters uses safe datasets, no adult pictures were used on the training.

You can watch the youtube tutorial on how I creatred this model here:

This LyCORIS has been trained after carefully selecting 89 images. All photos have been enhanced to make the most of them. I hope you like it!

Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

This is a lycors\locon you have to use the additional networks extension or the lycoris extension for this to work.

Enjoy the model for Min Min from ARMS. I personally didn't play ARMS at all I only know her from Super Smash Bros. in which she is a very controversial character.

I like to play her but i hate to play against her.

Keyword is Min_Min works well on 1.0 strength if you want to change the outfit lower the strength and if the clothes don't have the right color or if her mask doesn't show up increase it. I pruned only her hair and eye color but you might have to prompt them I had some pictures with black hair when I was generating the preview pictures and don't forget short hair.

As always if you need guidance look at the metadata but don't copy the metadata from the cherry blossom pictures these were RNG seeds and the model does seem to fail with these tags.

If you like the model leave a comment, like and rate it that would be appreciated.

Have fun generating.

Cinema Insomnia is a nationally syndicated American television series presented by horror host Mr. Lobo and OSI74.

"They're not bad movies -- just misunderstood."

This LyCORIS has been trained after carefully selecting 82 images. All photos have been enhanced to make the most of them. I hope you like it!

Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

Suggested Checkpoint - Chillout mix, basilKorea_v10, caulkinumv1SSeriesFor_s8

Suggested Lora weight - 1

DPM ++ 2M Karras, 20 steps,

Img 2 Img denoise 0.4

Rate or feedback me if you like the model

This LyCORIS has been trained after carefully selecting 102 images. All photos have been enhanced to make the most of them. I hope you like it!

Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

训练测试，质量比较一般，推荐权重0.6-0.8，基模Anything4.5

Training test, the quality is not very will, the recommended weight is 0.6-0.8, base model: Anything4.5

Thanks to @konyconi for the guide, and thanks to @navi_os for the dataset.

This LyCORIS has been trained after carefully selecting 63 images. All photos have been enhanced to make the most of them. I hope you like it!

Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

仅有单手模式

one hand only

可供参考的tag：

Tags for reference:

cleavage, bent over, areola slip

bikini,competition school swimsuit……

This LyCORIS has been trained after carefully selecting 77 images. All photos have been enhanced to make the most of them. I hope you like it!

Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

LyCORIS of Majestic Horse

This LyCORIS has been trained after carefully selecting 144 images. All photos have been enhanced to make the most of them. I hope you like it!

Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

skitter skitter skitter GLOP!

Honestly who knows, it makes gross robot spiders. Im working on a better version but it's kinda fun now so i'm putting it out there.

Should be back to work on the main cthulhuTECH model very shortly, nearly done with aircraft!

Prompting

Keyword is ctsp1d3r

Secondarys {mechanical|blooded|bone|digital|hitech|technical} you can use more than one

Stuff like, horror, grotesque, monstrous and mutation words like scaly skin, tentacles, etc all make cool beasties.

It is best for macro shots and has been trained with macro and close up as keywords

ps. Yes it will NSFW, is it suuuuuuuuuuper weird ? Yes.

If you like it, please review it or;

squeeze me some carrot juice - https://ko-fi.com/bunnyviking

>>Donate Coffee for Gtonero<<

if uses Conv2D requires need extensions
https://github.com/KohakuBlueleaf/a1111-sd-webui-locon
or
https://github.com/kohya-ss/sd-webui-additional-networks

Donate Coffee for Gtonero
Since training with Conv2D requires
https://github.com/KohakuBlueleaf/a1111-sd-webui-locon