This LyCORIS has been trained after carefully selecting 59 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.

LoHa,768768,anythingV5

AobaYuki/(denchi-project/),idol

发觉还有些空余时间临时制作的，有待改进。

服从度感人，之后会找个时间重新优化一下训练集。

This mod still needs improvement,I may improve it furture.

……我其实不知道这个应该叫啥

...I actually don't know what this should be called

使用方法就是写2girls或者3girls、4girls等等来控制屁股的数量，同时提高图片长度也会控制屁股的数量

Writing 2girls or 3girls, 4girls, etc. to control the number of butts, and increasing the length of the picture will also control the number of butts

on stomach和on back可以在一定程度上控制最下面那一位的身体朝向，它在2girls中比较适用，更多屁股存在时可能比较难以控制

on stomach and on back can control the orientation of the bottom girl to a certain extent, it is more suitable in 2girls, it may be difficult to control when there are more buttocks

如果你想使用人物lora来定向控制，我猜使用蒙版重绘就可以了

If you want to use the character lora for directional control, I guess inpainting with a mask will do the trick

至于控制衣服的部分我想不需要我多说了

As for the part of controlling the clothes, I don't think I need to say more

两个屁股和三个屁股比较好把控，更多数量需要看运气（我最高达到9个），例图中我展示了从2到7的一些图片，但是后面几张被审查了……

Two butts and three butts are easier to manage, more numbers are a matter of luck (I got up to 9), in the example picture I show some pictures from 2 to 7, but the last few are under review...

👑 GloryHole Lycoris👑

🔧Model base: AngrA_RealFlex_v02

___________

📝🌟 Trigger: (GloryHoleQuiron pose), <lyco:quiron_GloryHole_Lycoris_v1:1.0>, blowjob, penisuncensored, oral, fellatio, penis coming out of the hole in the wall ,

___________

📝 Variation: <lyco:quiron_GloryHole_Lycoris_v1:Y>, Y=0.4 to 1.0

📝 Denoising strength: 0.30 to 0.75

___________

🗒️💡Sample:

(best quality:1.1), (masterpiece:1), (realistic:1.3), 4k, (GloryHoleQuiron pose:1.2),

1girl, euro top model (perfect face, perfect eyes, perfect mouth, perfect body:1.2),

(blowjob:1.3),1boy, penis, huge penis, penis coming out of the hole in the wall ,

<lyco:quiron_GloryHole_Lycoris_v1c:0.8>

___________

🚀 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 )☕🌟

LoHa of Hishi Akebono (umamusume). Please install extension, a1111-sd-webui-lycoris, on WebUI. LyCORIS, not LoCon

After installing extension, put LoHa file in your stable-diffusion-webui/models/LyCORIS folder and write LyCORIS notation in prompt to applying LoHa. <lyco:your_loha_name:weight>

All of my works is free. If you'd like to support me, feel free to buy me a coffee. 👍

NOTICE

• After careful consideration, I decided not to produce LoRA of the characters below.

  • Otonashi Etsuko, Akasaka Misato, Hosoe Junko

• If anyone wants the characters above, I will consider making them.

• Except for the above, I will make the rest of the characters in Umamusume: Pretty Derby, unless there is a special notice.

Recommended options

• LoRA weight around 0.7

• Trigger words

  • Base

    • hishi akebono \(umamusume\)

  • Race :

    • hair ribbon, hair bobbles, mini hat, white chocker, striped dress, blue dress, white apron, buttons, badge, wrist cuffs, puffy short sleeves, waist apron, frilled apron, frilled dress, blue ribbon, skirt, white socks, white footwear, brown footwear, mismatched footwear, asymmetrical footwear, sneakers

  • Original Art :

    • hair ribbon, hair bobbles, blue bowtie, white jacket, double-breasted, buttons, long sleeves, blue skirt, pleated skirt, miniskirt, bandaid on knee, white socks, brown footwear, loafers

  • SSR Guts :

    • official alternate costume, hair bun, double bun, hair flower, pink flower, necklace, jewelry, collarbone, cleavage, bare shoulders, strapless dress, white dress, layered dress, wedding dress, frilled dress, elbow gloves, white gloves, dress flower, white bow, white thighhighs

  • Casual :

    • casual, hair ribbon, hair bobbles, necklace, heart, jewelry, white shirt, off-shoulder shirt, bare shoulders, center frills, white bow, long sleeves, suspenders, suspender skirt, brown skirt, plaid skirt, white socks, shoes

  • Tracen summer uniform

    • tracen school uniform, summer uniform, serafuku, puffy short sleeves, purple bowtie, horseshoe ornament, sailor collar, sailor shirt, purple shirt, white skirt, pleated skirt, frilled skirt, miniskirt, zettai ryouiki, white thighhighs, brown footwear, loafers

  • Tracen winter uniform

    • tracen school uniform, ear ornament, hairclip, winter uniform, puffy long sleeves, sailor collar, purple shirt, white bowtie, horseshoe ornament, purple skirt, pleated skirt, purple thighhighs, brown footwear, loafers

Use a model derived from or mixed with animefull model.

• Settings

  • Stable Diffusion WebUI by Automatic1111

  • DPM++ Series(SDE Karras, 2M Karras, etc.)

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Upscaler

      • Latent, Latent (nearest-exact), Latent (bicubic antialiased) or other Latent series.

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 92 images consist of

  • 4x low quality 33 images

  • 8x medium quality 36 images

  • 16x high quality 23 images

  • => 1 epoch = 788 images

• replaced character feature tags with hishi akebono \(umamusume\)

  • 1girl, horse girl, brown eyes, simple background, bangs, brown hair, long hair, hair between eyes, red eyes, pink eyes, solo, horse ears, animal ears, horse tail, official alternate costume, very long hair, medium hair, twintails, purple eyes, side ponytail, sidelocks

  • I think these tags are the essentials of Hishi Akebono or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

• other training settings is included in training data or metadata of LoRA file.

All uploaded images are generated by AOM3A1

And, special thanks to my best friend. DH.

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

First of all this is more of a proof of concept than a real model.

When i was playing through the game I thought: Is it possible to get the zora race working and can it handle that if I only have a handful of pictures to go with it ?" .. I managed to collect 22 pictures with some of them being screenshots from the game.

The model still works fine I have still some testing to do do consider this to be a preliminary upload. I will most likely upload another version in the near future when I either get better results from other epochs or when more images are being released and I have a larger dataset to play around with.

Keyword is Yona works on 1.0 strength if you miss some details you can adjust the weight.

Look at the metadata there are some tags that I think might be mandatory I'm still in the testing phase though. I think including horns and colored skin in the metadata might be mandatory if you dont want her to have nipples and a vagina you might have to put these in negatives or in positive depending on the pictures you generate.

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

Have fun generating.

LoHa,768768,anythingV5

KurokiMisa/(denchi_project/),cat ear headphones, purple shirt, military helmet, hat ornament, wristband, headphones around neck

发觉还有些空余时间临时制作的，有待改进。

This mod still needs improvement,I may improve it furture.

Who? :Laruna is the draph character in Granblue Fantasy. She is Narmaya's mother.

Weight :0.7~1.0

Trigger :larunamother,horns,hairclip,

Official Costumes Trigger : sleeveless,pink dress,larunacos,gloves,

Issue:

-The official costume "shawl" often appears other costumes.

　Using "shawl" for negative prompt might make it better...?

-hairpin,official cos is not stable.

-composition from the right is rare.

Welcome to Beautiful Pornstar Series Vol. 2! Let the beauty continues

--

If you love my work, you can buy me a coffee to help me stay awake and cope with life to create more lora and stuffs : ko-fi.com/tauron

Check vol.1 --> Tau's Beautiful Pornstar Vol. 1 [Fin] - 01. Gloria Sol | Stable Diffusion LyCORIS | Civitai

Best to use with my mix: Tauron Hybrid Mix - TauHybrid_Anime_Final | Stable Diffusion Checkpoint | Civitai

--

DISCLAIMER:

I'm doing this for fun and not for commercial motive, My main purpose is to compose a fictional character using AI with the beauty of all persons above, please support them if you love their works. i do not support any indecent images generated of all persons in this model, I do not support selling generated images of persons in this model. please refrain to what they have been posting online.

All persons in this models are 21+ and not a minor, please DO NOT post explicit nudity, suggestive content, or pornographic content online using this model. Refer to CivitAi TOS for more details.

===

Request or Suggestion Rule:

1. No request of minor!

2. The requested should have public data (no private)

3. Should have more than 30 good quality images

4. Since all requests or suggestion are free, I will still need to like your request, otherwise its a work not a hobby for me.

5. Be a decent human being and leave a review if your requests is up It takes you only 5 mins while it takes me hours, electricity, and efforts to do your requests.

6. If you don't get what you want, do not review bomb, there could be other creators models or lora that can get better results than mine, use theirs instead, and leave a good review on them too, support any other creators for their hardworks!

NOTE : I will keep updating more actresses in versions update

===

Usage Guide

--

A. Lora Weight

• 0.1 ~ 0.4 -- is to combine multiple faces or features, or if you experience overbaking

• 0.5 ~ 0.7 -- is a balanced weight to get still accurate feature but with more flexiblity

• 0.8 ~ 1.0 -- is to create accurate representation of the features.

B. CFG Scale

• Lower CFG Scale means AI gets more creative and less overbaking problems.

• Higher CFG Scale means results get very accurate with training sets and more prone to overbaking problems.

• Dynamic CFG Threshold, using this method allows you to bump up quality and achieve photorealistic easily, but with lycoris i found out recently that this method actually lower success rate, so only use when you are composing with a Lora / model that works fine with this method.

C. Sampling Steps, High Res Fix & Upscaling

• I like to use sampling steps at 50, with high res step at 20, or 1/2 of sampling steps.

• I find that 2M Karras works better and faster than SDE Karras. with ERSGAN 4x+ method + ERSGAN 4x for Upscaling. This method actually gives quite good result without being too sharp or too plasticky

• Upscaling - GPFGAN @0.25, Codeformer @0.3 & Codeformer Weight @0.15 seems to yield superior results to me.

--

Combining with Attribute LoRa / LyCoris

You may need to do your own testing, but using character at weight -0.2 below attribute lora weight yields superior results, if for some reason one features persists not showing or overbake, you may want to lower the weight of the overbaked feature Lora, and then use emphasis () to boost the specific attribute you want.

DDetailer / ADetailer Script

If subject face is small / far away use: DDetailer Script with bbox anime face with default settings except denoiser at 0.2-0.3. You can combine this LoRa with a lot of other Style / Attire / Pose / Acts LoRa, Also works well with character LyCoris.

How to install this script? I kinda forget how because its a bit complicated beacuse you need to install stuff from venv. but a quick google search will land you somewhere in tutorial page.

Enjoy.

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 them 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 supergirl outfit you just need to use the keyword "wearing supergirl_cosplay_outfit" maybe with a weight if necessary.

If you want a mix like supergirl|ballet or supergirl|bunnyyou should use both keywords, in this cases i have an important advice, play with weights! mostly when adding a particular color to the outfit. I make you an example, most of the training ballet dress images are white, most of the training supergirl outfits are blue, so if you write "(wearing ballet dress with tutu, wearing white supergirl_cosplay_outfit: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 supergirl_cosplay_outfit:1.3)". Opposite case if you want a blue supergirl ballet outfit.

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

真正的沙子雕像/制品艺术风格，一种使用沙子作为原材料的现代艺术，也可以出人物，具体效果可以参考图片

The real sand sculpture art style, a modern art that uses sand as a raw material, can also be out of character, specific effects can refer to the picture

材质模型的正确使用方式是图生图，效果贼拉棒

The correct way to use the material model is image to image, It works really very well

注意本材质一定要高修放大才可以有沙子的细节，并且不要使用面部修复

Note that this material must be hight fix to enlarge the details of the sand, and do not use face fix

1.0:

这个版本的沙子少一点

This version has a little less sand

1.1:

这个版本的沙子更多

This version has more sand

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 them 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.

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

Welcome to Insta Thot Series Vol. 2! Realize your thot fantasy here!

--

If you love my work, you can buy me a coffee to help me stay awake and cope with life to create more lora and stuffs : ko-fi.com/tauron

Check vol.1 --> Tau's Insta Thot Vol. 1 [Fin] - 01. [Req] @Imazhai | Stable Diffusion LyCORIS | Civitai

Best to use with my mix: Tauron Hybrid Mix - TauHybrid_Anime_Final | Stable Diffusion Checkpoint | Civitai

--

DISCLAIMER:

I'm doing this for fun and not for commercial motive, My main purpose is to compose a fictional character using AI with the beauty of all persons above, please support them if you love their works. i do not support any indecent images generated of all persons in this model, I do not support selling generated images of persons in this model. please refrain to what they have been posting online.

All persons in this models are 21+ and not a minor, please DO NOT post explicit nudity, suggestive content, or pornographic content online using this model. Refer to CivitAi TOS for more details.

===

Request or Suggestion Rule:

1. No request of minor!

2. The requested should have public data (no private)

3. Should have more than 30 good quality images

4. Since all requests or suggestion are free, I will still need to like your request, otherwise its a work not a hobby for me.

5. Be a decent human being and leave a review if your requests is up It takes you only 5 mins while it takes me hours, electricity, and efforts to do your requests.

6. If you don't get what you want, do not review bomb, there could be other creators models or lora that can get better results than mine, use theirs instead, and leave a good review on them too, support any other creators for their hardworks!

NOTE : I will keep updating more actresses in versions update

===

Usage Guide

--

A. Lora Weight

• 0.1 ~ 0.4 -- is to combine multiple faces or features, or if you experience overbaking

• 0.5 ~ 0.7 -- is a balanced weight to get still accurate feature but with more flexiblity

• 0.8 ~ 1.0 -- is to create accurate representation of the features.

B. CFG Scale

• Lower CFG Scale means AI gets more creative and less overbaking problems.

• Higher CFG Scale means results get very accurate with training sets and more prone to overbaking problems.

• Dynamic CFG Threshold, using this method allows you to bump up quality and achieve photorealistic easily, but with lycoris i found out recently that this method actually lower success rate, so only use when you are composing with a Lora / model that works fine with this method.

C. Sampling Steps, High Res Fix & Upscaling

• I like to use sampling steps at 50, with high res step at 20, or 1/2 of sampling steps.

• I find that 2M Karras works better and faster than SDE Karras. with ERSGAN 4x+ method + ERSGAN 4x for Upscaling. This method actually gives quite good result without being too sharp or too plasticky

• Upscaling - GPFGAN @0.25, Codeformer @0.3 & Codeformer Weight @0.15 seems to yield superior results to me.

--

Combining with Attribute LoRa / LyCoris

You may need to do your own testing, but using character at weight -0.2 below attribute lora weight yields superior results, if for some reason one features persists not showing or overbake, you may want to lower the weight of the overbaked feature Lora, and then use emphasis () to boost the specific attribute you want.

DDetailer / ADetailer Script

If subject face is small / far away use: DDetailer Script with bbox anime face with default settings except denoiser at 0.2-0.3. You can combine this LoRa with a lot of other Style / Attire / Pose / Acts LoRa, Also works well with character LyCoris.

How to install this script? I kinda forget how because its a bit complicated beacuse you need to install stuff from venv. but a quick google search will land you somewhere in tutorial page.

Enjoy.

My first LyCORIS, which works on GoddessV2.

Make sure to install a1111-sd-webui-lycoris extension and put the model into the LyCORIS folder (NOT the LoRA folder), and use "lyco:" instead of "lora:"

Note that it will NOT work on SD 1.x models, and I cannot guarantee if it works in other SD 2.1 models (since I haven't tested).

训练测试，质量一般，推荐权重：0.7，基础模型：AnythingV4.5。

如果帽子出不来可以试试加入：A long vertical pointed hat with a black color and golden decorations

触发词：KuonHat 也可以起到效果。

有些模型会丢失一些特征，加入这些tag以重新获得特征。

long hair, white hair, blue eyes

加入这些tag以做出其标准动作。

hand up, index finger raised,

使用时，请多加质量词，强烈推荐使用单轴512-768以内的分辨率生成之后使用ControlNet-Tile配合Ultimate SD Upscale放大生成结果。

Training test,quality is not very will, recommended weight: 0.7, base model: AnythingV4.5.

If the hat does not come out, you can try adding:

A long vertical pointed hat with a black color and golden decorations
The trigger word: KuonHat can also work.

Some models may lose some features, add these tags to regain features.

long hair, white hair, blue eyes

Add these tags to make Kuon's standard actions.

hand up, index finger raised,

When using, please add more quality words, strongly recommend using 512-768 resolution within a single axis to generate and then use ControlNet-Tile with Ultimate SD Upscale to enlarge the generated result.

Locon-Lycoris Locon-Lycoris

Locon-Lycoris Locon-Lycoris

Do not re-distribute this Locon Data.

Private Trading and Public Trading of this Locon Data is NOT allowed.

Please use this Locon Data at your own risk. I can not be held liable or responsible for any damages caused by using this data.”

This Locon Data may not be used for any works which contain negative content toward the original product, individual, &/or group.

This Locon data is made for personal use only. Please do not post images that expose the lower half of the body or suggest sexual depictions on sites that can be viewed by an unspecified number of people.　Please refrain from posts that require "18禁" and "nsfw" tags.

AI images created from this Locon data are not permitted to be sold in any form, whether digital or printed. Please do not use this LoRA data for commercial purposes. A subscription type like PIXIV FANBOX is also useless. It is also not good to post a URL link of a paid service on a page with a healthy AI image to guide it.

I'm not releasing Locon data to fill your wallet

Use the Lora at 0.4-0.7 weight for optimal results.

Required prompt: hasegawa langa\(sk8\), blue eyes, blue hair

close up

I used google translate. I think there may be mistranslations, but please accept that I am serious about writing in Japanese below.

原作者様・その他関係者様にご迷惑にならない範囲でご使用ください。

※ここで言うLocon データは出力された生成結果も含む。

この Locon データを再配布しないでください。

この Loconデータのプライベート取引およびパブリック取引はやめてください

この Locon データはご自身の責任でご利用ください。 このデータの使用によって生じたいかなる損害についても、私は一切の責任を負いません。

この Loconデータは、オリジナルの製品、個人、および/またはグループに対する否定的、攻撃的な内容を含む作品に使用することはしないで下さい。

このLoconデータは個人使用を目的として作成されています。 不特定多数の人が閲覧できるサイトに、下半身の露出や性的描写、性転換や異性衣装を着衣を示唆する画像の投稿はご遠慮ください。簡単に言うなら「18禁」「nsfw」タグが必要な投稿はご遠慮ください。

ローカル環境（あなたのPC内で完結するのであれば私は干渉せず）で楽しむ分には問題ありません。※それでも健全なファンアートとしてSNS（PIXV、Twitter、deviantartなど）に載せたい場合はタグやコメントに「AI Illustrator」「AI Generate」とAIで作成されたのだと分かるように記載をお願いします。

この Loconデータから作成された AI 画像は、デジタル、印刷を問わず、いかなる形式でも販売することはできません。この Locon データを商用目的（金銭が一円でも発生する事）に使用しないでください。PIXIV FANBOX のようなサブスクリプション型、広告収入の素材としても一切認めません。また、このLoconで作成された画像が掲載されているページに有料サービスの URL リンクを貼って誘導するのもやめて下さい。

私はあなたの財布の中身を増やすために LoRA データをリリースしているわけではありません

生成した画像に少し手を加えた。PNG情報を消したり、jpegに変換したから証拠は残らない、証拠が残らなければ大丈夫という考えは通じません！

目覚ましい発展を続けるAI技術に法整備やサイトの運営方針が急に変わる可能性もあります。ファン活動の一環としてLoconを公開しています。法律、版権元様のルールに従います。

Locon-Lycoris Locon-Lycoris

Locon-Lycoris Locon-Lycoris

Do not re-distribute this Locon Data.

Private Trading and Public Trading of this Locon Data is NOT allowed.

Please use this Locon Data at your own risk. I can not be held liable or responsible for any damages caused by using this data.”

This Locon Data may not be used for any works which contain negative content toward the original product, individual, &/or group.

This Locon data is made for personal use only. Please do not post images that expose the lower half of the body or suggest sexual depictions on sites that can be viewed by an unspecified number of people.　Please refrain from posts that require "18禁" and "nsfw" tags.

AI images created from this Locon data are not permitted to be sold in any form, whether digital or printed. Please do not use this Locon data for commercial purposes. A subscription type like PIXIV FANBOX is also useless. It is also not good to post a URL link of a paid service on a page with a healthy AI image to guide it.

I'm not releasing Locon data to fill your wallet

Use the Lora at 0.4-0.7 weight for optimal results.※Hoodie logo not so good,Yellow is preferred. If you have a desired color, good luck! !

Required prompt: kyan reki\(sk8\), red hair, hair between eyes, messy hair【headband,】

close up

I used google translate. I think there may be mistranslations, but please accept that I am serious about writing in Japanese below.

原作者様・その他関係者様にご迷惑にならない範囲でご使用ください。

※ここで言うLocon データは出力された生成結果も含む。

この Locon データを再配布しないでください。

この Loconデータのプライベート取引およびパブリック取引はやめてください

この Locon データはご自身の責任でご利用ください。 このデータの使用によって生じたいかなる損害についても、私は一切の責任を負いません。

この Loconデータは、オリジナルの製品、個人、および/またはグループに対する否定的、攻撃的な内容を含む作品に使用することはしないで下さい。

このLoconデータは個人使用を目的として作成されています。 不特定多数の人が閲覧できるサイトに、下半身の露出や性的描写、性転換や異性衣装を着衣を示唆する画像の投稿はご遠慮ください。簡単に言うなら「18禁」「nsfw」タグが必要な投稿はご遠慮ください。

ローカル環境（あなたのPC内で完結するのであれば私は干渉せず）で楽しむ分には問題ありません。※それでも健全なファンアートとしてSNS（PIXV、Twitter、deviantartなど）に載せたい場合はタグやコメントに「AI Illustrator」「AI Generate」とAIで作成されたのだと分かるように記載をお願いします。

この Loconデータから作成された AI 画像は、デジタル、印刷を問わず、いかなる形式でも販売することはできません。この Locon データを商用目的（金銭が一円でも発生する事）に使用しないでください。PIXIV FANBOX のようなサブスクリプション型、広告収入の素材としても一切認めません。また、このLoconで作成された画像が掲載されているページに有料サービスの URL リンクを貼って誘導するのもやめて下さい。

私はあなたの財布の中身を増やすために LoRA データをリリースしているわけではありません

生成した画像に少し手を加えた。PNG情報を消したり、jpegに変換したから証拠は残らない、証拠が残らなければ大丈夫という考えは通じません！

目覚ましい発展を続けるAI技術に法整備やサイトの運営方針が急に変わる可能性もあります。ファン活動の一環としてLoconを公開しています。法律、版権元様のルールに従います。

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.

This is a Lycoris model, and you need to install a1111-sd-webui-locon extension to use this model.

v1.5 NO Trigger word, v1.0 Trigger words: oil painting, realistic

Suggest weight: around 0.9 or lower.

CFG scale: 5.5-7

Check point recommendation: sunshinemix_sunlightmixPruned, Anything V4.5, CounterfeitV30_v30

Based off Hidari's work on Fire Emblem Echoes. Trained on character art from the game.

LoHa of Seeking the Pearl (umamusume). Please install extension, a1111-sd-webui-lycoris, on WebUI. LyCORIS, not LoCon

After installing extension, put LoHa file in your stable-diffusion-webui/models/LyCORIS folder and write LyCORIS notation in prompt to applying LoHa. <lyco:your_loha_name:weight>

All of my works is free. If you'd like to support me, feel free to buy me a coffee. 👍

NOTICE

• After careful consideration, I decided not to produce LoRA of the characters below.

  • Otonashi Etsuko, Akasaka Misato, Hosoe Junko

• If anyone wants the characters above, I will consider making them.

• Except for the above, I will make the rest of the characters in Umamusume: Pretty Derby, unless there is a special notice.

Recommended options

• LoRA weight around 0.7

• Trigger words

  • Base

    • seeking the pearl \(umamusume\)

  • Original Art :

    • ear ornament, eyewear on head, sunglasses, red-framed eyewear, bracelet, jewelry, necklace, black gloves, off shoulder, bare shoulders, detached sleeves, long sleeves, red jacket, star \(symbol\), belt, pleated skirt, white skirt, sandals, high heels

  • SSR Speed :

    • official alternate costume, ear ornament, eyewear on head, sunglasses, red-framed eyewear, military uniform, white gloves, long sleeves, white jacket, white skirt, pleated skirt

  • Casual :

    • casual, ear ornament, eyewear on head, sunglasses, red-framed eyewear, jewelry, long sleeves, vertical stripes, striped shirt, bracelet, belt, clothes around waist, black skirt, pencil skirt, high heels, red footwear

  • Tracen summer uniform

    • tracen school uniform, summer uniform, serafuku, puffy short sleeves, purple bowtie, horseshoe ornament, sailor collar, sailor shirt, purple shirt, white skirt, pleated skirt, frilled skirt, miniskirt, zettai ryouiki, white thighhighs, brown footwear, loafers

  • Tracen winter uniform

    • tracen school uniform, ear ornament, hairclip, winter uniform, puffy long sleeves, sailor collar, purple shirt, white bowtie, horseshoe ornament, purple skirt, pleated skirt, purple thighhighs, brown footwear, loafers

Use a model derived from or mixed with animefull model.

• Settings

  • Stable Diffusion WebUI by Automatic1111

  • DPM++ Series(SDE Karras, 2M Karras, etc.)

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Upscaler

      • Latent, Latent (nearest-exact), Latent (bicubic antialiased) or other Latent series.

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 43 images consist of

  • 4x low quality 13 images

  • 8x medium quality 14 images

  • 16x high quality 16 images

  • => 1 epoch = 420 images

• replaced character feature tags with seeking the pearl \(umamusume\)

  • 1girl, animal ears, solo, purple eyes, horse ears, grey eyes, mole under eye, black hair, horse tail, very long hair, green eyes, two-tone hair, lips, hair between eyes, blue eyes, brown hair, official alternate costume, bangs, horse girl, parted lips, long hair, red lips, multicolored hair, simple background

  • I think these tags are the essentials of Seeking the Pearl or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

• other training settings is included in training data or metadata of LoRA file.

All uploaded images are generated by AOM3A1

And, special thanks to my best friend. DH.

I created this model as a request :)

The dataset was huge, so there is a lot of combinations. Imagination its your limit XD

Enjoy!

Note: The dataset used in this model its safe, no adult images were used.

LoHa of Tsurumaru Tsuyoshi (umamusume). Please install extension, a1111-sd-webui-lycoris, on WebUI. LyCORIS, not LoCon

After installing extension, put LoHa file in your stable-diffusion-webui/models/LyCORIS folder and write LyCORIS notation in prompt to applying LoHa. <lyco:your_loha_name:weight>

All of my works is free. If you'd like to support me, feel free to buy me a coffee. 👍

NOTICE

• After careful consideration, I decided not to produce LoRA of the characters below.

  • Otonashi Etsuko, Akasaka Misato, Hosoe Junko

• If anyone wants the characters above, I will consider making them.

• Except for the above, I will make the rest of the characters in Umamusume: Pretty Derby, unless there is a special notice.

Recommended options

• LoRA weight around 0.7

• Trigger words

  • Base

    • tsurumaru tsuyoshi \(umamusume\)

  • Original Art :

    • ear ornament, hair ornament, hairclip, wide sleeves, long sleeves, detached sleeves, bare shoulders, cleavage cutout, clothing cutout, strapless, sleeveless coat, purple coat, white shorts, single thigh strap, brown footwear, toeless footwear, sandals

  • Tracen summer uniform

    • tracen school uniform, summer uniform, serafuku, puffy short sleeves, purple bowtie, horseshoe ornament, sailor collar, sailor shirt, purple shirt, white skirt, pleated skirt, frilled skirt, miniskirt, zettai ryouiki, white thighhighs, brown footwear, loafers

  • Tracen winter uniform

    • tracen school uniform, ear ornament, hairclip, winter uniform, puffy long sleeves, sailor collar, purple shirt, white bowtie, horseshoe ornament, purple skirt, pleated skirt, purple thighhighs, brown footwear, loafers

Use a model derived from or mixed with animefull model.

• Settings

  • Stable Diffusion WebUI by Automatic1111

  • DPM++ Series(SDE Karras, 2M Karras, etc.)

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Upscaler

      • Latent, Latent (nearest-exact), Latent (bicubic antialiased) or other Latent series.

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 93 images consist of

  • 4x low quality 40 images

  • 8x medium quality 26 images

  • 16x high quality 27 images

  • => 1 epoch = 800 images

• replaced character feature tags with tsurumaru tsuyoshi \(umamusume\)

  • 1girl, parted bangs, medium hair, two-tone hair, blonde hair, simple background, horse tail, horse girl, horse ears, white hair, grey eyes, long hair, alternate hair length, brown hair, purple eyes, blank eyes, green hair, brown eyes, multicolored hair, alternate hairstyle, streaked hair, red eyes, solo, short hair, orange hair, animal ears, pink eyes, bangs

  • I think these tags are the essentials of Tsurumaru Tsuyoshi or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

• other training settings is included in training data or metadata of LoRA file.

All uploaded images are generated by AOM3A1

And, special thanks to my best friend. DH.

trigger word : hooters

if you like my stuff pls consider supporting me on Kofi

accepting Commissions as well...

Beelzebub, Sin of Gluttony from the Seven Mortal Sins (Sin: Nanatsu no Taizai).

LoHa trained on NAI (anime). Requires the LyCORIS plugin.

Full body images seem kind of biased when using the outfits though. Ill update if I find a way to fix it. However, I'm not quite sure if its really possible due to lack of source material.

Trigger words:

• beelzebub - For character. Helpful tags: "twintails, wings"

• beelzebuboutfit - For main outfit. Add "apron, gloves, frills, white thighhighs" for best results.

Experimental (undertrained) trigger words:

• beelzebubempress - For alternate outfit from mobile game. Add "cat ears, tail, black thighhighs, drill hair, blue flower" for best results.

• beelzebubgastro - For alternate outfit from mobile game. Add "japanese clothes, fundoshi, sarashi, thigh strap, jingle bell" for best results.

• beelzebubdress - For alternate outfit from mobile game. Add "chinese clothes, navel cutout, detached sleeves, white thighhighs, jewelry" for best results.

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.

By the nature of the original art style, this is going to bias heavily towards making chrome ladies. It's a hard life, folks.

Notes

• This is definitely not perfect.

• It's rather sensitive and needs experimentation. For example, in Sorayama's real work, there are occasionally actual flesh faces peeking through the chrome. If that's your intention, try lowering the strength (about 0.25? 0.4?).

Loha for BB from Fate. Includes her standard outfit, nurse uniform and all three different outfits from her swimsuit form. Weight 1 is ok. Use different weights for the tags, if it does not work well.

Tags for the different outfits:

Standard: bb,1girl,og_ver, pointing_stick (pointing_stick is optional)

Nurse: bb,1girl,nrs,glasses,syringe (syringe and glasses are optional)

First ascension of the swimsuit version: bb,1girl,fst_asc_lulu

Second ascension of the swimsuit version: bb,1girl,sec_asc_lulu,tan,red_eyes (tan and red_eyes are optional)

Final ascension of the swimsuit version: bb,1girl,fin_asc_lulu,black_coat,tan,red_eyes (black_coat,tan and red_eyes are optional)

This LyCORIS has been trained after carefully selecting 68 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.

If you'd like to support me and do more:
https://ko-fi.com/edgonkofi - commissions open!
If you're looking for a >>LoRA Making Tutorial<<

Lycoris Hoodies

As technology progresses, one must make experiment and try new things.

Try out LoCon, it's nice and it makes great hoodies!

Download this LoCon, and be sure to give me your feedback!

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

Model for the character Hisaishi Kanade from the Anime Hibike Euphonium.

The character does not appear until the last movie though and has a really nasty personality but I personally like her for that.

Keyword ist Hisaishi_Kanade works great on 1.0 you can adjust it if you need more or less accuracy. I pruned her hair color and eye color but if you want it 100% accurate you should prompt them as well. Doesn't get her uniform 100% maybe if you adjust the strength but is very flexible with her outfit in general

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

Have fun generating.

This LyCORIS has been trained after carefully selecting 55 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.

Hiya Papaya! I 100% don't have a type.
Here's what costumes work(more or less)

• Uniform: psgkneesocks, demon_uniform, necktie

• Swimsuit: psgkneesocks, demon_swimsuit, glasses,bare_shoulders, navel

• Demon outift: demon_outfit, halterneck, bare_shoulders, glasses, skirt, cleavage, navel

The demon outfit is the hardest to get right especially if you want to wings. Might retrain for better wings if I decided to revisit this.

Helpful terms: single_horn, glasses

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

Enjoy the model from Viola from Pokemon

When i collected the dataset for Greta I had one image from her and i really liked her design so I choose to collect a dataset for her as well, once i started training i noticed that there are already models for her but I still want to upload my model

Maybe it's smaller than the other or you like lycoris models more or for whatever reason I appreciate you using my model

Keyword ist Viola works on 1.0 strength i think if you want to change her outfit set it a bit lower.

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

Have fun generating

警告：不是很稳定，会导致肢体错误

WARNING: Not very stable, can cause limb errors

很多时候肢体问题很大，但是我暂时解决不了

Many times the physical problem is very serious, but I can't solve it temporarily

我比较建议生成横图

I recommend generating a horizontal graph

参考prompt：

reference prompt:

1girl,1boy,lying on person, girl on top,hug

2girls,lying on person,yuri

👑 Blowjob Sideview Lycoris 👑

🔧Model base: AngrA_RealFlex_v02

________________

📝🌟 Trigger: BlowjobSideviewQuiron pose, <lyco:quiron_BlowjobSideview_Lycoris_v1:1.0>, blowjob, side view, oral

________________

📝 Variation: <lyco:quiron_BlowjobSideview_Lycoris_v1:1.0>, Y=0.4 to 1.0

📝 Denoising strength: 0.30 to 0.75

________________

🗒️💡Sample:

BlowjobSideviewQuiron pose, <lyco:quiron_BlowjobSideview_Lycoris_v1:0.7>

________________

🚀 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 )☕🌟

其实吧，这个是一个style loha来着的，只不过发现这个loha貌似还原人物还原的挺好就变成character loha了.....（其实还是我想试试看loha的参数选什么比较好误打误撞练出来的一个loha）没有触发词，但是要一些额词来辅助否则出不来老婆这个人物（等我考完试就再搞搞loha训练画风和人物什么的，感觉效果确实是比lora好什么的挺离谱的）

1280步，62张图，batchsize2，repeat4，参数如下，tagger1.4可信度0.3，全标

0.55-0.6,anythingv5,LoHa.

akagi_rin\(denchi-project\),star hair ornament, twin braids

uniform

bodysuit, mechanical ears, vambrace, pauldrons, but this still requires further imorovement.

一些碎碎念 2023/5/28

作为一部所谓宅文化毁灭系的片子，电池少女在剧本上是平庸的，但并不是差劲的，尤其是跟四月的魔法少女毁灭者相比，更是展现了作为一部原创动画剧情上能够平稳落地的含金量。

然而这个片作画上的浓浓万策尽气息让它几乎不太能成为又一部小众佳作……并且让这个模型诞生的十分艰难！

我实际上一个多月前就训练了这个角色的Lora，并于近日做好了三个女主角的全部Lora，但都在服装的服从度上均表现的很微妙，我原本想打算直接发布，但经朋友提醒，这两天又用数据集赶出了这么个loha出来。

LoHa的确在角色的学习上非常优秀，比方说它甚至学到了我没扣干净的素材痕迹，以至于我不得不放弃我用恋活截图得到的一些素材。。不过在这个LoHa中你仍然能看到这一点，这显然是个过拟合的版本，但我暂时要忙于别的事无法测试，所以先发布这个模型，，

除了关键词的这些，这个角色还有护膝，创可贴，鱼形拉链等特征我没有还原，因为动画一个劲的拍面部特写和上半身导致我缺乏素材……