LyliaEngine

Pony Diffusion V6 is a versatile SDXL finetune capable of producing stunning SFW and NSFW visuals of various anthro, feral, or humanoids species and their interactions based on simple natural language prompts. CHECK "ABOUT THIS VERSION" ON THE RIGHT IF YOU ARE NOT ON "V6" FOR IMPORTANT INFORMATION. Please join our Discord Server to support development of new versions of this model and get access to free SD bot and check out more examples of this model capabilities on our prompt sharing website or follow the author on Twitter. Important information Make sure you load this model with clip skip 2 (or -2 in some software), otherwise you will be getting low quality blobs. This model supports a wide array of styles and aesthetics but provides an opinionated default prompt template that allows generation of high quality samples with no negative prompt and otherwise default settings score9, score8up, score7up, score6up, score5up, score4up, just describe what you want, tag1, tag2 (previous Pony Diffusion models used a simpler score9 quality modifier, the longer version of V6 XL version is a training issue that was too late to correct during training, you can still use score9 but it has a much weaker effect compared to full string. You can learn more about these tags here). The model is designed to not need negative prompts in most cases and does not need other quality modifiers like "hd", "masterpiece", etc... Other special data selection tags include, 'sourcepony', 'sourcefurry', 'sourcecartoon' and 'sourceanime' and ratings of 'ratingsafe', 'ratingquestionable' and 'ratingexplicit'. This model is capable of recognizing many popular and obscure characters and series. If you are looking specifically for pony style, I recommend using one of the two following templates `anthro/feral pony, rest of the prompt` or `sourcepony, rest of the prompt`. This model is trained on combination of natural language prompts and tags and is capable of understanding both, so describing intended result using normal language works in most cases, although you can add some tags after the main prompt to boost them. Using Euler a with 25 steps and resolution of 1024px is recommended although model generally can do most supported SDXL resolution. This model will sometimes generate pseudo signatures that are hard to remove even with negative prompts, this is unfortunately a training issue that would be corrected in future models. If that's an issue for you I suggest trying V5.5 or inpainting. Iceman for helping to procure necessary training resources PSAI Server Subscribers for supporting the project costs PSAI Server Moderators for being vigilant and managing the community The model has been trained on ~2.6M images aesthetically ranked based on authors personal preferences, with roughly 1:1 ratio between anime/cartoon/furry/pony datasets and 1:1 ratio between safe/questionable/explicit ratings. About 50% of all images has been captioned with high quality detailed captions, which results in very strong natural language capabilities. All images has been trained with both captions (when available) and tags, artists' names have been removed and source data has been filtered based on our Opt-in/Opt-out program. Any explicit content involving underage characters has been filtered out. License This model is licensed under a modified Fair AI Public License 1.0-SD (https://freedevproject.org/faipl-1.0-sd/) license. The following modifications have been added to Fair AI Public License: You are not permitted to run inference of this model on websites or applications allowing any form of monetization (paid inference, faster tiers, etc.). This applies to any derivative models or model merges. If you want to use this model commercially, please reach us at [email protected]. Explicit permission for commercial inference has been granted to CivitAi and Hugging Face. You should use `None` to trigger the image generation. Weights for this model are available in Safetensors format.

All Disney Princess XL LoRA Model - Ralph Breaks the Internet This LoRA model is trained on screen capture images featuring beloved Disney princesses from the movie Ralph Breaks the Internet, including Rapunzel, Snow White, Ariel, Aurora, Belle, Cinderella, Elsa, Anna, Jasmine, Mulan, Merida, Tiana, Moana, and Pocahontas. It's designed to be an All-in-One Model for your creative endeavors. Check out SD1.5 Version Here: https://civitai.com/models/202866/all-princesses-from-disneys-ralph-2-ralph-breaks-the-internet-or-all-in-one-lora-model Additionally, this model includes representations of other characters like Vanellope, Shank, Raya, Namaari, Asha, and Rapunzel with short hair. While these extras are included, the quality may vary. Direct Name Usage: You can generate any princess by directly prompting their names, such as Anna, Ariel, Aurora, Belle, Cinderella, Elsa, Asha, Jasmine, Merida, Moana, Mulan, Namaari, Pocahontas, Rapunzel, Raya, Shank, Snow White, Tiana, Vanellope. More Character in v2 including: Alice, Chel,Esmerada, Jane Porter, Kida, Megera,Mirabel Madrigal, Isabela Madrigal, Dolores Madrigal and some secret character! LoRA Scale: Optimal LoRA scales range between 0.6-0.7 for close-up character generation and 0.3-0.5 for full-body generation. Combining these scales using the ADetailer plugin can yield enhanced results. Ensure an adequate number of sampling steps for better output. Base Models: Although various base models can function, models with animation styles like DreamShaper or RealCartoon are preferred for optimal performance. You should use `None` to trigger the image generation. Weights for this model are available in Safetensors format.

Should work well on any XL/Illustrious/NoobaAI Checkpoints! Clothes Details: created and merged Loras that adds more intricate details to clothing texture. Skin Details: created and merged Loras that add a little more details to skin. Should be more noticeable on realistic Checkpoints. Armor and Weapons Details:: created and merged Loras that adds detailed engraving and patterns. Better material quality. The more Loras you use the lower the Lora weight should be! A detail tweaker/booster inspired by the work of many awesome creators like ShedTheSkin's Detail Slider LoRA, w4r10ck's Detail Tweaker XL, and many, MANY, others. This gives an overhaul improvement to color, lighting, shadows and adds more details. This Lora was made to be used on with Smooth Mix and Smooth Embeddings, but of course, they should work well with any other Checkpoints/Embeddings/Loras. Just try out your favorites and see how it turn out! You should use `None` to trigger the image generation.

Just my junky models I use to fix up my MJ/Niji images I merge a bit...maybe a bit too much 😈 and sometimes I mix models that I end up liking, so I thought I would share, Will they work for you? I do not know... but good luck 👍 Yet again another custom models I have been using, really like the look...so I share :p This one is...I don't know tbh, not really trad anime, its more on the illustration side (but it could do semi real stuff too), have tried loras on this one...but I don't know how it behaves when stacking too many of them...so good luck if you use a butt load 😏 WARNING!!!! : when using loras be sure to adjust....you have been warned 😈 ADETAILER is highly advised, please use it - https://github.com/Bing-su/adetailer (If you can't use adetailer I advised to to not use this model.) If you must, use inpaint to fix issues like messed up hands. if you need help with prompting just take the parameters from my images, I will always share my prompts :D need more help, just hit me up, always happy to help. I have been using both models to help with my artwork, both offer something different, similar? Maybe but I have been alternating with them. I usually just make stuff that fits my tastes, it may not be for everyone and that's okay, but hey if you find it helpful or you end up enjoying it, that's all that matters right? 😊 tamtamx1332 for the amazing block merger comfyui workflow ♥ and for the base model 🙏 sulph for making this: https://civitai.com/articles/6183/merge-block-weights (Thank you 😭) most of the models are merged to oblivion (using the block merge method) so I end up forgetting the parent model (original). I will remember to write down whenever I take a parent model and add it to credits. Permissions on this model: if you ever use it to merge and upload the model, I just ask you allow merge share and free download. I believe in open source and freedom, I want the public to be allowed to better things with out restrictions and I want to allow others to be able to share creations. AI art is a blessing. You should use `None` to trigger the image generation. Weights for this model are available in Safetensors format.

u asked abt it. 😊 Now I present to u: Animij ~ retro soft cored. 2 different styles ~ moaarrrrrr possibilities?! 😋 For a long time I asked myself why merge an anime model ~ when there are a wealth of great models that serve this genre? Ur voice is important to me, so I took the plunge. U know me, I like things different, and so Animij spawns as an anime style that may not suit everyone's taste ~ but maybe urs? Tasty for more? Maybe u wanna play around with version 2? In Animij, details whisper of emotion ~ a stroke of storytelling. Eyes shimmer with depth, reflecting unspoken dreams. Light and shadow dance across soft gradients, giving each face a warmth that feels almost alive. The magic lies in the small things: the delicate blush of a cheek, the glint of light on a ring, the way colors breathe with life. Nothing is static~ every detail wanna invite u: wanna come deeper? making each piece feel like a moment suspended between fantasy and reality. ✨ Wanna look closer? The story is in the details. ♥ setting examples V2 V2 can be used without -prompt ♥ setting examples V1 masterpiece, highquality, highres, flatcolors, gradient, masterpiece, highquality, highres, sharpfocus, detailedeyes, retro, 3D, realistic, realisticbody, sharpeyes, worstquality, badquality, poorlydetailed, young, nude, 2d, sketch I am happy if u enjoy and stay curious abt ur results ~♪~ If u like what I do consider leaving me a like or a comment, much appreciated and helps me getting a feeling for tinkering ~♥ You should use `None` to trigger the image generation. Weights for this model are available in Safetensors format.

This is a lora that helps realistic image generation, you can use my checkpoint to get the results I posted below (yomama 2.5D [Illustrious] [Pony] - Illustrious v1.0 | Illustrious Checkpoint | Civitai). This checkpoint can generate 2.5D results pretty well, but the face tends to be more anime/cartoonish, it can hardly generate realistic image; with this filter, we can shift the balance to a more realistic result. This lora works on other ckpts too, but the results may vary, you may need to play with the weight a bit. e.g. Illustrious-XL - v0.1 | Illustrious Checkpoint | Civitai e.g. NoobAI-XL (NAI-XL) - Epsilon-pred 1.1-Version | Stable Diffusion XL Checkpoint | Civitai p.s. Big thanks to creator @FallenIncursio for giving me advice all the time. 😀 You should use `realistic` to trigger the image generation.

if i may introduce ~ posILustrij ~ illustrij's soft sister. She is posi', dynamic and simply soft. A real anime lady who likes to cuddle with semi. How intense? Belonging to ur own taste ~♥ By playn around u'll maybe feel, she is not real, she is not purely drawn, that's how she spawns, soft but clean, that's how she is, the in-between, like anime and semi kissed. ~♥ general prompting: "masterpiece, best quality, high resolution, very aesthetic, absurdres, newest, professional, high-quality, hyper-realistic, portrait of a woman, looking at viewer, detailed eyes, realistic body," either I did not use a negative prompt or e.g. the following: "(lowres:1.2), (worst quality:1.4), (low quality:1.4), (bad anatomy:1.4), multiple views, jpeg artifacts, artist name, young," Feel free to copy or for sure use ur own description ☺ or if u'r interested in more ➡️ https://civitai.com/articles/10509/il-portreij-step-by-step For higher distance I would recommend using an ADetailer. I am happy if u enjoy and stay curious abt ur results ~♥ You should use `None` to trigger the image generation. Weights for this model are available in Safetensors format.

Valid for illustrious type model only. > pony ver. < No unauthorized reproduction. It affects my motivation. 😭It would be encouraging if you could press the like button.😭 Designed to distance certain prompts from "mature female". Negative values emphasize “mature female” while Negative values emphasize “mature female”. The “reverse” version reverses the effect and can be used with online generators that cannot use negative numbers. You should use `mature female` to trigger the image generation. Weights for this model are available in Safetensors format.

You should use `4ki4k4neillu` to trigger the image generation. Weights for this model are available in Safetensors format.

If you appreciate my work and want to see more updates and models from me in the future, please like, follow, share the art you create with my models, and tip me some Buzz, your support is what motivates me to continue this work, thank you! UPDATE July 2025: Well I said I wouldn&#39;t do it but here I am, you all have shown me such love and support I wanted to give back, so I thought what better way to return the love than using the skills I have gained to refine my most popular model, the difference is very subtle, I want to be very careful to maintain the integrity of this model as much as possible, but I figured I have learned enough by now to be able to achieve a task like that- When I first made this model all I knew was basic merges, and I rather chaotically slapped together what I thought would work nicely together, each iteration becoming vastly different than the previous version, now my methods are more surgical and precise, so I&#39;m going to try and refine this model as best as I can, this first refinement focused on touching up the faces and expressions (which is the constructive criticism I seemed to get most often for version 4), but it also seems to have slightly cleaned up the background noise, and added a very light improvement to skin tone&#x2F;texture, lighting, and shadow. this is a test to see if this kind of thing is something you guys even want, if it isn&#39;t as well received or better than the previous version I&#39;ll remove this refinement because I&#39;m still a bit worried to cover up the version 4 gallery with a new update when that gallery is so incredible. I hope you guys enjoy this, I would love to hear your feedback, let me know if you like the polished version better or if you want me to keep up with these refinements, and what you think I could and should refine within this model for the future, and please put her through her paces and test her out for me, make the gallery go Brrrrrr. All showcase comparison images for this finetune created with DMD2 4 step Lora, 8 steps, Euler a, CFG 1, with adetailer. UPDATE: If you like this model please try out my latest realistic illustrious mix WAI x REAL linked in the suggested resources below, in my opinion it is a far superior realistic Illustrious model, and you can consider it a spiritual successor, and the closest you will get to an update on this model, this showcase is far too good to hide behind any version updates. Thank you everyone, and especially DevrabbitAI for your outstanding contributions to this models showcase, I hope to see the WAI x REAL gallery one day surpass this one. I&#39;ve been obsessed with Illustrious lately so I decided to make my own realistic mix, I&#39;m very happy with the result and I hope you will enjoy her as much as I am, this is a merge of Illustrious, RillusmRealistic, PornmasterPro, and RedcraftCADS Positive prompt: DSLR,realistic,cosplay photo, (very awa,masterpiece,best quality,year 2024,newest,absurdres,highres:1), realistic,skin texture,facial features, photography shot with Canon EOS 5D Mark IV , realistic, RAW, best quality, masterpiece, amazing quality, skin details, Negative: lowres,worst aesthetic,bad quality,worst quality,bad anatomy,sketch,jpeg artifacts,scan artifacts,lossy-lossless,ugly,poorly drawn, (illustration,3d,2d,painting,cartoons,sketch,anime:1.4),toon,oil painting \(medium\),flat color,outline,smooth skin, (blurry:1.4) (some users have experienced fried purple generations with the recommended prompts, if you have this issue try changing the 1.4&#39;s in negative to 1&#39;s) Please enjoy and create responsibly everyone. I&#39;m looking forward to your feedback, and seeing what you create with my girl. You should use `None` to trigger the image generation.

I’m hosting my model on SocialDiff.net—a project I’m developing in my spare time. ADetailer is fully integrated, making your experience even smoother. If you enjoy the model, don’t forget to leave a like and a review—your feedback helps improve the project! Also, support the talented creators publishing their work with this model by giving them some well-deserved appreciation. 🌟 Digging my way to the Illustrious Checkpoint. 🌟 • Exceptional Attention to Detail: Every aspect is thoughtfully refined. • Beautifully Detailed Bodies: Impressive proportions and structure. • Perfectly Crafted Hands: No awkward poses or inconsistencies. • Illustrious Vision for Semi-Realistic Results: A balance between realism and artistic finesse. masterpiece, best quality, amazing quality, very aesthetic, detailed eyes, perfect eyes, realistic eyes • All images are polished using ADetailer (faceyolov8s) for flawless face results. You should use `None` to trigger the image generation. Weights for this model are available in Safetensors format.

¡Presentamos nuestro modelo LoRA de sprites para Ragnarok Online en Citivai! 🎮✨ Con más de 190 imágenes de alta calidad, es perfecto para los fans y creadores que buscan llevar su creatividad al siguiente nivel. ⚔️ Únete y colabora con otros apasionados de Ragnarok Online en Citivai. ¡Juntos podemos hacer crecer esta colección épica! You should use `Black filled oval eyes` to trigger the image generation. You should use `Ragnarok online character` to trigger the image generation. You should use `ROSPRITE` to trigger the image generation. Weights for this model are available in Safetensors format.

Pony World of Warcraft Dwarf - By HailoKnight Hello friends! This Lora was created as a collaboration, Game Characters , over at the AIPub Discord. Be sure to drop by and join our amazing community! Please be sure to check out the other entries in this collaboration: Squeej: Lord Admiral Jaina Proudmoore from World of Warcraft Flashfiring: Skaven, Children of the Great Horned Rat You should use the trigger words dwarf (For best results) The LoRA weight I have been using is 0.85, but feel free to try and experiment! I have used ComfyUI when generated pictures and the settings im using are: Sampler: dpmpp2m Scheduler: Karras Steps: 30 For A1111 enjoyers, I would use: Sampler: DPM++ 2M Karras Steps: 30 &quot;score9, score8up, score7up, score6up, score5up, score4up, (A dwarf from the world of warcraft, beautiful green eyes, he is in full green armor and holding an axe while standing in a magical forest. The background features colorfull magic mushrooms:1.2)&quot; First image is with my LoRA and second is without my LoRA, same prompt and model. You should use `dwarf` to trigger the image generation. Weights for this model are available in Safetensors format.

This lora made from mixing two type of style from Midjourney and Flux ( Midjourney v6 and Flux pro ). Realistic, hyper-realistic, hyperrealistic, portrait, closeup, semi-realistic Note : The lora also works well with realistic models, It increases the details and also adds some additional effects like a detailer lora In the end it&#39;s just an experiment and you may encounter some problems unfortunately, thank you for your time You should use `Realistic` to trigger the image generation. You should use `hyper-realistic` to trigger the image generation. You should use `hyperrealistic` to trigger the image generation. You should use `portrait` to trigger the image generation. You should use `closeup` to trigger the image generation. You should use `semi-realistic` to trigger the image generation. Weights for this model are available in Safetensors format.