AI Models

Explore our collection of AI models for image generation

ComfyUI Model

OpenAI Model

I try to pin images that stand out / i personally like every 1-3 weeks So if i unpin yours doesnt mean that i dont like it anymore just trying to keep a fresh rotation of images without completly filling the start page with pins <3 Join the Discord for questions / sharing pictures :) : https://discord.gg/gcJqAKQ5Af If you enjoy my models and want to support me, you can buy me a coffee! (Since my patreon got disabled like a year ago) 🛠️ Recommended Settings: • Resolution: best results with 832x1216 //960x1440// 1024x1024 // 1024x1536 • CFG Scale: 5–8 • Sampling Method: Euler a or DPM++ 2M Karras • Steps: 20–50

+2

hey guys, ☺ as u know I rlly love comic-anime-hybrid based checkpoints and digital art rendering with a touch of 3D on its way to semirealism. So I've thought it would be time to test Illustrious. This is how I wanna present to u ~ Illustrij ~ my first merged Illustrious checkpoint :) 💌 v20 NEW ♥ Hey friends, I’m happy to introduce v20 ☺ Your feedback really counts ~ and you can feel it here. v20 leans more toward v18 in look, attitude, and overall style: very Illustrij, just a little more grown up ^^ This new version also includes one of my private LoRAs, based on my artificial handwriting. I hope you enjoy exploring with Illustrij as much as I do ✨ with ♥ Reij 💎 | Setting (option examples): • Euler a • steps 30 • CFG 7 💎 | Hires Setting (option examples): • 832x1216 💎 Hires x 1,5 • Denoising strength 0,4 • Hires steps 10 • Hires CFG Scale 4 • 832x1536 💎 Hires x 1,5 • Denoising strength 0,2 • Hires steps 10 • Hires CFG Scale 2 💎| Quality Tags (option examples): | + | masterpiece, best quality, | + | masterpiece, best quality, ultra-detailed, 8k resolution, high dynamic range, absurdres, stunningly, intricate details, sharp focus, detailed eyes, cinematic color grading, high-resolution texture | ~ | worst quality, bad quality, 🎶 about v19 sooo… Illustrij update? I’d say she’s matured^^ “grown”? more like years instead of months :P ~Illustrij spawns: more NSFW focus, more adult vibe ~ that’s important to me. perfection? I say: creativity! Illustrij ~ a mirror of my own excursions, adventures, and discoveries in the AI world^^ of course with lots of love and curiosity ♥ artistic, sensual eroticism… a part of me. spark-seeker? yes? and definitely a fan of shadow-play ^^ illustrij reflects this shift in my artistic sense ~ and still remains completely illustrij: stubborn, playful, full of surprises, and always an invitation to dream, explore, and be creative ^^ I’m super excited and curious to see what you create ^^ feel free to tell me if you enjoy this adventure. I’m happy about every heart traveling along with Illustrij :) with ♥ Reij | ♥ | Specs: Euler a / DPM++ 2M Karras· 30 Steps · 832×1216 · CFG 4-7 · Hires ×1.5 · SFW & NSFW | ♥ | quality tags option examples: | + | masterpiece, best quality, ultra-detailed, 8k resolution, high dynamic range, absurdres, stunningly beautiful, intricate details, sharp focus, detailed eyes, cinematic color grading, high-resolution texture, photorealistic, | ~ | worst_quality, bad_quality, poorly_detailed, 💌 NEWS for ♥ hearts that grew with the v6 line 🎶 Illustrij BTTR https://civitai.com/models/2060715?modelVersionId=2331934 is the natural continuation of the v6 era ~ a soft fusion of past and present. Back to the roots ~ the handwriting of what once was, now glowing again. for those who still feel the light between present lines. ✨ 🎶 about v18 Ur feedback is important to me ☺ Why no update for so long? Sometimes u take a step back to make the next leap. | ♥ | a little finer | ♥ | a touch more restraint & therefore more on point | ♥ | less shine, more shimmer | ♥ | one more dance step towards my hidden aim ~ ADetailer optional ~ less must, more can | ♥ | Specs: DPM++ 2M Karras· 30 Steps · 832×1216 · CFG 4 · Hires ×1.526 · SFW & NSFW | ♥ | quality tags option examples: | + | masterpiece, best quality, | ~ | worst quality, bad quality, young, I'm happy if u feel comfortable with the new version ♥ ~♥ Reij 🎶 about v17 recommended settings examples: • 30 steps • Euler a • Clip skip 2 • no extra VAE (VAE included) • tags / _tags • resolutions 832x1216 +++ +prompt: masterpiece, best quality, ultra-detailed, 8k resolution, high dynamic range, absurdres, stunningly beautiful, intricate details, sharp focus, detailed eyes, cinematic color grading, high-resolution texture, Small mini upgrader for face focus images with hand motions, gives images a more focused touch regardless of the prompt: photorealistic portrait, nails, -prompt: (worst quality:2), (low quality:2), (normal quality:2), bad anatomy, bad proportions, poorly drawn face, poorly drawn hands, missing fingers, extra limbs, blurry, pixelated, distorted, lowres, jpeg artifacts, watermark, signature, text, (deformed:1.5), (bad hands:1.3), overexposed, underexposed, censored, mutated, extra fingers, cloned face, bad eyesI am happy if u enjoy and stay curious abt ur new adventures with Illustrij with ♥️ Reij about v16 Back to the Core, Forward in Form 🔹 Illustrij 16 is a fusion ~ a conscious pause between old and new. She remains true to herself and recognizes her original reflection more than ever. What seems like a return is also progress: a rediscovery of the origin with a new perspective. The semi-realistic base with a subtle anime touch remains, but now with a touch more video game aesthetic ~ stylized but tangible, inspired by character designs that balance between playable figure and digital fantasy. The look is more oriented towards the origins ~ before the big 2.5D shift ~ and picks up the iconic face shape of previous versions: familiar, rounder, clearer. This time, the special focus is on the realistic reaction of the body: skin that responds to movement, light and shadow that create depth and presence. SFW and NSFW are still supported - but with more emphasis on the body through credible lighting and play of form. v16 wraps itself in a cocoon of “Back to the Roots”, interwoven with a sweet-cheeky duality ~ sweet but cheeky, like at the beginning. From this inner reconciliation a new self emerges: a balance between nostalgia and artistic development. recommended settings examples: • 30 steps • Euler a • Clip skip 2 • no extra VAE (VAE included) • tags / _tags • resolutions 832x1216 +++ +prompt: masterpiece, best quality, high quality, absurdres, very aesthetic, 8k, depth of field, subject focus, face_focus, detailed eyes, -prompt: lowres, bad anatomy, deformed face, watermark, logo, optional for the “tiny extra” glossy look 🫦 +prompt: masterpiece, best quality, absurdres, gradient, face_focus, detailed eyes, very awa, nail polish, glossy skin, (+ optional: realistic skin)-prompt: bad quality, worst quality, lowres, jpeg artifacts, bad anatomy, signature, watermark, censored, I am happy if u enjoy and stay curious abt ur new adventures with Illustrij with ♥️ Reij about v15 🔹 Illustrij 15 keeping a semirealistic look by spawning with hyperrealistic, near-photographic skin textures, a clean airbrush finish, and cinematic lighting. The look blends digital editorial aesthetics with anime-inspired elegance. Her clothing is form-flattering with textile-realistic sheen – v15 is polished, stylized, and ready to bring ur imagination to life :) situational: this version tends to be a bit more body-forward, so if that's not ur style, adding “nudity” to the negative prompt helps tone it down. Ur feedback matters ~ that’s why this update brings more "depth". I am happy if u have fun exploring and can’t wait to see what u create! ~♥ with ♥️ Reij recommended settings examples: • 30 steps • Euler a, DPM++ 2M • Clip skip 2 • no extra VAE (VAE included) • tags / _tags • resolutions 832x1216 (◄• showcase images) & 1040x1510 +prompt: high_quality, highres, detailed_eyes, detailed, masterpiece, best quality, absurdres, 8k, HDR, face_focus, alternatively: masterpiece, high quality, very_awa, newest, absurdres, highres, depth_of_field, realistic_skin,-prompt: lowres, bad anatomy, deformed face, alternatively: poorly_detailed, jpeg_artifacts, worst_quality, bad_quality, lowres, bad anatomy, deformed face, glossy_skinabout v14 🔹 Illustrij v14 ~ she moves with us While preparing the showcase, she caught me all over again ~ and that’s what I love most about Illustrij. We meet her with or without expectations ~ it doesn’t matter. She surprises, reveals new shades of herself, and stirs that quiet curiosity that pulls us deeper. She dances with the prompt ~ playful, bold, sensual. Not loud, but vividly alive. Her look is familiar, yet this time she feels more tangible than ever ~ softer fills, intentional lighting, and a glow that embraces instead of overwhelming. Her clothing? Almost like skin ~ clinging to her with every movement, emotion, breath. As if it wants to sway with her, feel with her, live with her. Illustrij v14 doesn’t live in the frame. She lives the frame. 🔸 Showcase Note: All showcase images are rendered at 1040 × 1510 ~ because she just wanted to show a little more of herself. A bit more room, a bit more motion. That’s so Illustrij: more pixels, more presence, more of that signature sway. 🩰✨ If she twirled her way into ur heart too ~ she’d love a ♥️. And so would I. ✨🩰 I am happy if u enjoy and stay curious abt ur results ~♥ with ♥️ Reij recommended settings examples: 30 stepsEuler aClip skip 2CFG 7no extra VAE (VAE included)tags / _tagsresolutions 1040 x 1510+prompt: masterpiece, detailed_eyes, high_quality, best_quality, highres, absurdres, 8k, subject_focus, depth_of_field, -prompt: poorly_detailed, jpeg_artifacts, worst_quality, bad_quality, about v13 🖤 Tiny update ~ big heart ✨ Hey friends~ although I wasn't planning on doing a revision so quickly, a quiet spark inside me made me want to update Illustrij with a lil more soul ~ especially in her hands, her gestures, her presence. She's still herself, just a touch more expressive. ✨ At the same time, on-site generation has brought new challenges ~ and one of them is staying true to her unique face without needing facefixes. That’s why I’ve been merging and refining more often lately. It's not just tinkering ~ it’s me listening closely to her, shaping her gently, and trying to bring her essence through without compromise. It’s a small change, but it carries a lot of love. Maybe it’s just a subtle shift… or maybe it’s the detail someone was waiting for. Either way, I couldn’t keep it to myself ♡ But I didn't want to just overwrite her former self. So: The current version is now available for free download as a small thank u and souvenir ~ and in a few days the updated version will gently take her place in the on-site generation, assuming we make it through the auction. Thank u so much for being here through all her forms ♡ Every detail, every evolution ~ it means the world to share it with u. Let’s keep creating ~ together, gently, beautifully. With 💖 ~ Reij All showcase images are in on site resolution 832 x 1216 with hires x1,526 ~ without Adetailer recommended settings examples: 30 stepsEuler aClip skip 2CFG 7no extra VAE (VAE included)tags / _tagsresolutions 1024*1024++prompt: high_quality, highres, detailed_eyes, beautiful, detailed, masterpiece, best quality, absurdres, 8k, HDR, face_focus, ☺ I decided to showcase her in a new vibe: 🎨 Retro Pop Art Grunge. ☺ To get the showcase image style feel free to add: retro, grunge, popart+prompt: worst_quality, bad_quality, poorly_detailed, about v12 ✨ Hey friends~ Back again with a new version of my precious comic-anime hybrid: Illustrij v12 🌸 U know how much I love that digital art rendering that leans into soft 3D, with a touch of lightplay and emotion. And with v12, she’s blooming more than ever ~ with more depth, warmth and expression. She’s not just a model ~ she’s a little soul. A vibe. A mood. A presence. I don’t just build tools ~ I shape characters. And she, like the others, feels like her. 🖤 This version has a stronger sense for light, character design and stylistic harmony ~ especially suited for natural and studio-style shots. I focused the previews more on my current fav: soft retro-grunge aesthetics 🧃🌾 but she’s still versatile like always! (The other styles are all still possible, give it a try~) If she speaks to u ~ a like, a comment or a creation means the world. Thank u for being here ♡ All showcase images are in on site resolution 832 x 1216 with hires x1,526 ~ without Adetailer about v11 hey guys, ☺ the current version places more focus on light, prompting, Illustrij character design and depth. recommendet settings examples: 30 stepsEuler aClip skip 2CFG 7no extra VAE (VAE included)_tagsresolutions 1024*1024+Locally u can of course also use a resolution around 1040x1510 ☺ +prompt: high_quality, highres, detailed_eyes, beautiful, detailed, masterpiece, best quality, absurdres, 8k, HDR, face_focus, masterpiece, high_quality, highres, depth_of_field, subject_focus, 8K, (masterpiece:1.3), (best_quality:1.3), (ultra_detailed:1.2), (official_art), (absurdres), intricate_details, refined_textures, volumetric_lighting, perfect_shadow, realistic_depth, hyperfocus, (depth_of_field:1.2), detailed_eyes,-prompt: worst_quality, bad_quality, poorly_detailed, extra_fingers, malformed_hands, deformed_feet, blurry, youngworst_quality, bad_quality, poorly_detailed,about v10 Hey guys, ☺ over the last few days I've continued to work on Illustrij, spawned v10 with softer skin, new expression ~ to create a bit more liveliness. Even with the latest changes, Illustrij isn't perfect - I'm continuing to work on making it possible to create images without Adetailer. If the distance is further away, I would currently recommend Facefix or Adetailer. The new onsite generation presents me with new challenges ~ so I continue to tinker diligently and still experimenting, v10 is more oriented towards v8 again with an inclusion of v9. The newest version is more balanced in the mix between semi-realism and anime elements. recommendet settings examples: Euler aClip skip 2CFG 5-7no extra VAE (VAE included)_tagsresolutions 1024*1024+☺ mostly I'm using the following resolutions: 832*1216 & 1040*1510 or 720*1280 & 900*1600 ☺ for hands and feet u are welcome to use the following quality tags: +prompt: detailed_hands, elegant_fingers, beautiful_feet-prompt: extra_fingers, malformed_hands, deformed_feet,example quality prompt recommendation: +prompt masterpiece, high_quality, highres, sharp_focus, detailed_eyes,-prompt: worst_quality, bad_quality, poorly_detailed, extra_fingers, malformed_hands, deformed_feet, blurry,Have fun being creative ☺ I am happy if u enjoy and stay curious abt ur results ~♥ a little comparison V7 to V10: about v9 Tinkered around a bit with my own models and Loras back to a more balanced anime-semireal style and added a little NSFW touch. ☺ hentai & NSFW With hentai and NSFW images, the model sometimes has its own idea the less ur pre-prompting ~ maybe u would like to get involved? These two posts offer a little foretaste: https://civitai.com/posts/14082774 https://civitai.com/posts/14081639 more animebased: https://civitai.com/posts/14088147 Since Illustrij is based on combining elements of semirealism and anime, the newer versions spawn with more depth and a 2.5 D feeling. Through promptings we can determine the direction a bit, whether more semi-realism or anime ☺ entirely according to ur own taste. For more anime I recommend the following settings: recommendet resolution 1024*1024 / 832*1216 / 720*1280 recommendet settings examples Anime+ Clip skip 2Euler local: 15-25 stepsCFG 7-10+prompt (masterpiece, high_quality, highres, anime_style, flat_colors, gradient), retro_artstyle 1980 \(style\), 2D, sketch, - prompt worst_quality, bad_quality, poorly_detailed, young, realistic, 3D, semirealrecommendet settings examples Semirealism+ Clip skip 2Euler aon site: 50 steps / local: 25+CFG 4 (2-7)+prompt (masterpiece, high_quality, highres, flat_colors, gradient),best_quality, absurdres, ultra-detailed, highly_aesthetic, highly_detailed_eyes, depth_of_field, subject_focus,(masterpiece, high_quality, highres, flat_colors, gradient), ultra-detailed, highly_aesthetic, highly_detailed_eyes, depth_of_field, subject_focus,-prompt example worst_quality, bad_quality, poorly_detailed,bad_anatomy, low_quality, poor_lighting, out_of_focus, misaligned, poorly_drawn, cluttered_background, unnatural_pose, disproportionate, unattractive_expression, soft_shading, clean_lines, no_background, In both variations, Illustrij remains a mix of semi-realism and anime, with different emphasis. ☺ In summary, the more steps and the lower the CFG, the more semi-realism, the fewer steps, and the higher the CFG, the more anime-touched. I am happy if u enjoy and stay curious abt ur results ♥ about v8 V8 is an experiment to integrate some of my Loras into Illustrij ~ without triggers of course ;) Let me know if u like the direction and how we can develop Illustrij further. I also tried to ensure that the model's style remained true to itself while leaving out the negative prompt. V8 can create furry aswell. I'm looking forward to ur results and hope u have a lot of fun being creative. recommended settings: Euler a / Euler25-30 stepsCFG 5-7Clip skip 2_ might wanna be ur friend for tags☺ Showcase resolution: 720*1280. I would recommend using ADetailer for further distance. +prompt example masterpiece,best quality,amazing quality,very aesthetic,absurdres,newest,detailed eyes,-prompt example u can leave emptyor worst_quality, bad_quality,a little comparison of V7 and V8 about v7 V7 takes a big step back to the roots and yet somewhere completely different ~ quest: keeping the balance . . . focus on painting a 2.5D lightning effect by drawing the Illustrij faces. I'm curious to see whether u like it and look forward to ur creations ~♥ Please let me know if u like the new direction ☺ recommended settings: 720*1280 / Euler a / CFG 7 / steps 25+ / Clip skip 2 🐰 furry optional about v6 The new versions spawns with a touch of 2.5 D remdering ~ with a flat-shaded anime style, but subtle 3D lighting and depth, making images look almost tangible. I hope you like it and stay tuned for your results ~♥ • showcase resolution: 1040 x 1510 prompt examples: Prompting is basically similar to the previous versions. After playn around with and testing old quality prompts ~ at the moment I prefer to use "masterpiece, best quality, amazing quality, very aesthetic, absurd, newest, detailed eyes" or "masterpiece, best quality," for +prompt and "worst quality, bad quality," for -prompt. Of course these are just examples, u can vary them to suit ur own taste. about v5 hey guys, ☺ the gap between v4 and v5 was only a few days ago, but I've been "tinkering around" a bit and didn't want to keep it from you. Trying to work on getting the quality closer to hires without hires - but there are still a few steps to go ^^ For people who have good hardware and can generally use a lot of Loras etc. at the same time the difference might not be that big, but I feel for the people who have small graphics cards :I The quality isn't perfect, but I've been working on it a bit ~ soft filling, clean renderings ~ the plan was to create an own style, subtly painted ~ anime-based, but still kissed by reality. I hope you like the merge and I'm really looking forward to your results ♥Belonging to the style u prefer u can switch the promptings 4 sure by ur own taste. In general, I mostly use "masterpiece,best quality,amazing quality,very aesthetic,absurdres,newest,detailed eyes" for general quality prompting, and besides depending on whether I want more anime or anime semi 3D touched I am writing "A professional,high-quality,hyper-realistic portrait of a woman, 3D" for the semianime 3D version and "A professional,high-quality portrait of a woman" for anime e.g. or as a conclusion for semi-anime: "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," and neg.: "(lowres:1.2), (worst quality:1.4), (low quality:1.4), (bad anatomy:1.4), multiple views, jpeg artifacts, artist name, censored, young, 2D" Sometimes I either leave out the negatives completely or write, for example, "lowres, worst "aesthetic, bad quality, worst quality, bad anatomy, sketch, jpeg artifacts, ugly, young, soft rendering, face marks" depending on whether I use Euler or Euler a as sampler ^^ Usually I structure my prompts based on this: • subject description: who or what can be seen? (what does she/it look like? hair, eyes, clothing, description of the scene), • light and shadow description, • which angle, perspective, motion? quality and • detailed description or the quality at the very beginning and then the rest of the structure :) If u're tasty for more? Here is a lil insight into the prompt structure: https://civitai.com/articles/10509/il-portreij-step-by-step about v4 hey guys, ☺ for V4 I changed the recipe a little to improve the quality a tiny bit and also to expand the options for prompts. about v3 style closer to V1 anime based with a lil qualy upgrade ☺ Showcase images without ADetailer about v2 a lil more 3D touched without using 3D in promptings by writing "realistic" the style will become more into semirealism and without its closer to a semianime based style furry optional a lil quality upgrade, just a lil ^.^ 🌶️ sidefact for NSFW fans: viewer can interact with character by describing what hands can do (see sample images in the showcase) about v1 If u like the style in the showcase images, u can add high-resolution, realistic, 3D, detailed skin, detailed eyes, shadows, dark light, eyelashes, upper body, cheeky, posing, holographic color, to ur prompt. Without it the style becomes more anime based. At further distance I would recommend using ADetailer. Today I can't say yet in which direction the merger is developing, but as a pony and flux fan, I can only say that I love the color intensity of Illustrious. I recently tested it and got a taste for it, maybe you'll like it too ☺ I am happy if u enjoy and stay curious abt ur results ♥ a small comparison from V1 to V5 ⬆️ masterpiece, best quality, 1girl, face, portrait, looking at viewer, closed mouth, close up worst quality, bad quality ~~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~* I hope it gives a little impression of the development ☺

+3

Follow me on X/Twitter for exclusive content: https://x.com/Citron_Legacy 😁 Style Couture Craze returns on ZImageTurbo!

+3

Recommended Settings (v1 - Z-Image Turbo): • WORKFLOW IS IN THE SHOWCASE IMAGES. You have to use the workflow. If you use the LoRA with a simple workflow you: • Either get shitty overtrained plastic images • Or you barely change the style of turbo model and they might still look shitty If you use it with my workflow, the seed variety is phenomenal. There is a very very less chance that you get the same-ish composition and lighting twice Prompt adherence took a little hit (maybe its the workflow to blame). It is okay for me because it still generates the images I want. I loose a bit of adherence to get variety. If you dont like it then you can try other workflow's or reduce the weight of the LoRA or use another LoRA Some prompts change the image way too much when you compare with Turbo but some prompts not so much. You can start with weight 0.7-0.9 for "Model with LoRA High Weight" node and 0.4-0.5 for "Model with LoRA Low Weight" node. The last stage (stage 3), disable the LoRA so you can bring back the skin texture - the image barely changes here because we use a low denoise. You can use the upscaler that I am using but keep in mind that its slow compared to others If you want to use a character LoRA along with this LoRA, you have to enable the character LoRA at stages 2 and 3. In stage 2, you can unbypass the character lora loader node. In stage 3, you can use "Model with Character LoRA Only" node Recommended Settings (v1 - Qwen Image): • Workflow is in the showcase images • My recommendation is to use the "EasyCache" node and let it skip 2 steps. It good with very little quality loss • I never tested it without lightning lora so I don't know how many steps to use or the CFG. You have to experiment yourself • 880x1184 or 1328x1776 works best Recommended Settings (v6 - Flux Dev): • Distilled CFG Scale: 3.5 • Sampling method and Schedule type: DEIS with DDIM • Steps: 20 • Resolution: 896x1152 • Hires fix model: 4x_NMKD-Superscale-SP_178000_G • Steps: 10 • Denoise: 0.3 • Upscale by: 1.5 • Lora Weight: 0.8. You have to experiment based on your prompts

+3

A LoRA inspired by one of my favourite styles of hyperdetailed colored pencil illustrations. This style captures the intricate shading and fine textures, enhanced by the subtle depth and warmth of colored pencil strokes. Ideal for creating striking portraits with expressive details, this style brings out the natural beauty of facial features, elaborate clothing, and delicate embellishments. It blends artistic sophistication with a touch of classical elegance. The keywords are ArsMJStyle, Colored pencil hyperdetailed realism. V2 Update: I simply love this LoRA , so I've been steadily extending the dataset with more diverse images , while trying to maintain the Original Concept. It is more flexible than V1 , but with just a slight sacrifice for the Style Strenght. Use V1 for stronger Effects or V2 for greater flexibility . MidJourney V7 Update: A complete rework of the LoRA’s dataset using MidJourney V7 for improved versatility and future expansion. The original aesthetic has been preserved as best as I can after switching models, with a shift toward more realistic rendering.

+3

The module already has VAE. Unless you have specific coloring requirements, it is recommended to use the VAE inside the module. Currently, personally recommend NoobAI (1.9, 2.6,2.7) . If the FP32 upload fails multiple times, stop attempting to upload. Suggested: Txt2img 720p+ with Hires fix. Upscaler recommended Remacri setting to 1080p+. • Prompt Prefix:(If you are using a suffix, add BREAK in front.) masterpiece,best quality,perfect quality,absurdres,newest,very aesthetic• Negative Prompt: shiny skin, shiny clothing, worst quality, low quality, lowres, signature, username, logo, bad hands, mutated handsSuggested Resources: Eyeful, adetailer, remacri, Face_finder Simplified instructions for the standard guidelines of NoobAI Character and artist tags Prompts achieve reliable performance when applied to NOOBAI CSV files with the Danbooru+e621 dataset exceeding 300 images in the official records. ( NoobAI huggingface csv). The illustrious checkpoint pre-release beta version for testing must be used in accordance with the illustrious XL standard usage guidelines.This illustrious checkpoint is not stable, and the background sometimes disappears Those pony checkpoint fused by indigoXL, zavychromaxl v9.0 and forrealxl v0.5. Its usage is basically the same as the indigo model. You can refer to the introduction of the indigo model. Here's the link.

+3

快速上手 | Quick Start 这是什么？ | What is this? • People's Works是一个实验性的微调模型系列，最初基于一个由Pony V6 XL生成的图片构成的数据集。这个模型的数据集由数千张AI社区用户发布的图片和作者使用AI合成的图片构成，经过人工筛选、编辑、修改和标注后用于训练。除此之外还有一个超过2000张由照片、游戏CG和3d渲染图像构成的辅助数据集，用于额外补充知识。 • People’s Works is an experimental fine-tuned model series, originally based on a dataset composed of images generated by Pony V6 XL. The dataset consists of several thousand images published by AI community users, along with images synthesized by the author using AI. These images were manually curated, edited, modified, and annotated before being used for training. In addition, there is an auxiliary dataset of over 2,000 images composed of photographs, game CGs, and 3D renders, used to provide supplementary knowledge. 模型功能 | Model features • 本系列模型主要功能是帮助基础模型在不使用画师串、较少的质量提示词的条件下获得相对稳定的风格化图像，为提示词节约token空间。 • 使用人工挑选和标注的数据集强化训练正面和负面审美提示词。 • 模型针对flat color、realistic等特定风格肌理进行了强化。 • 对人物的年龄、族裔、肤质等特征的精细化控制。 • 使用了较高的训练分辨率，使基础模型在高清修复时表现更好。 • 通过手工修复图片的方式降低了模型的某些细节瑕疵出现的概率。 • The primary function of this series of model is to help the basemodel generate relatively stable, stylized images, without artist keywords or long quality tags, freeing up token space for prompts. • By using manually selected and annotated datasets, the model strengthens both positive and negative aesthetic tags training. • The model includes targeted enhancements for specific visual textures and styles, such as flat color and realistic. • It enables finer control over character attributes, including age, ethnicity, and skin texture. • A higher training resolution is used, improving the basemodel’s performance during high-res upscaling. • By manually editing the images, the likelihood of certain flaws appearing in the model’s outputs has been reduced. 使用方法 | Usage positive: masterpiece, best quality, very aestheticnegative: low quality, displeasing 更新记录 | Change log v9 • 更改了系列名称。自这个版本起，训练集中来自Pony v6 XL的图片已经在所有AIGC内容中占比不足1/3。随着越来越多的新模型出现，我有计划在明年将这个系列拓展到其他模型上。为了避免未来用户使用上的混乱和误解，这个系列从这个版本起更改命名。 The series name has been changed. Starting from this version, images sourced from Pony v6 XL make up less than one third of the training data across all AIGC content. As increasing number of new models are emerging, I plan to expand this series to other models next year. To avoid potential confusion and misunderstanding for users in the future, the series name has been changed starting from this version. • 这个版本的训练方式是直接训练LoCon，而非训练Checkpoint后再抽取LoRA。模型相较前一个版本效果更强。 This version is trained directly as a LoCon, rather than training a checkpoint first and then extracting a LoRA. Compared to the previous versions, the model delivers stronger effects. • v9全系列使用1536分辨率的训练集。现在使用这个Lora生成图片时支持单边768-1536的分辨率，使用高清修复时也可以尝试更高的denoise参数了。 All v9 models use a 1536-resolution training set. When generating images with this LoRA, single-side resolutions from 768 to 1536 are now supported. When using high-res fix, you can also try higher denoise values. • 对训练图片调色。现在模型在没有指定色彩时更倾向于生成暖色调的图片，并且色彩的饱和度略微提高。较暗的场景明暗对比更大了。 Color adjustments were applied to images. When no specific color is specified, the model now tends to produce warmer tones, with slightly increased saturation. Darker scenes also have stronger contrast between light and shadow. • 之前版本的数据集中，人物鼻子的画法不统一。出于作者本人的兴趣，手工修改了其中约300幅图片，并暂时排除了约200幅来不及修改的图片。现在人物的鼻子有鼻翼了。 In earlier versions of dataset, nose depiction was inconsistent. Out of personal interest, the author manually modified around 300 images and temporarily excluded about 200 images that could not be edited in time. Characters’ noses now have nose wings. • 删除了旧版本中数百张过时的低质量训练数据。 Hundreds of outdated, low-quality images from older versions of dataset have been removed. • 添加了一个新的实验性数据集： 使用真人相片作为引导，现在你可以使用以下年龄和族裔标签了： A new experimental dataset has been added. Using real photographs as guidance, you can now use the following age and ethnicity tags: child, teenage, adult, matureCaucasian, Asian, Indian, African• 我的标签设计优先选择Danbooru数据集中已经存在的标签，尽管其中很多只有很少量的数据，在原版模型中几乎无法触发。重新启用了已经被删除的danbooru词条Caucasian和teenage，增设了adult和African两个标签。此外，loli和shota因为其文化背景中强烈的性暗示倾向，这两个词条被完全替换，根据具体情况分流入child和teenage。 My tag design prioritizes labels that already exist in the Danbooru dataset, even though many of them had very limited data and are therefore almost impossible to trigger in the base models. The previously removed Danbooru tags Caucasian and teenage have been re-enabled, and two new tags, adult and African, have been added. Additionally, due to the strong sexual connotations of loli and shota in their cultural context, these tags have been completely replaced and redistributed into child and teenage depending on the situation. v8 肌理更新：强化了以下tag的学习： Texture Update: The following tags have been reinforced in training: realistic, photorealistic, flat color,shiny skin, matte skin, shiny hair,请注意，在danbooru数据集中有很多个用于描述“照片”和“接近照片的风格”的tag。我在训练集中统一标注这些图片为“photorealistic”。但是使用danbooru训练集训练的SDXL模型大多并不能很好地画出写实图像，因此“photorealistic”只建议在较小权重下用于改变画面的肌理。“realistic”可以在高权重下正常运作。 Please note that Danbooru dataset contains multiple tags to describe "photo" or "photo-like styles". I’ve tagged all such images as “photorealistic” in dataset. However, most SDXL models trained on the Danbooru dataset do not render realistic images well. “photorealistic” is only recommended at low weight, where it can help adjust texture rather than create realism images. The “realistic” tag can work properly at higher weight. v8版本简介： • Pony: People's Works （ppw）是一个实验性的微调模型系列，数据集有约85%是收集自CivitAI上用户发表的AI生成图片。早期ppw的数据集最初建立在由pony v6生成图片的基础上，因此本系列模型生成的图片也带有pony diffusion的特征。 • 本系列模型使用标准Danbooru标签，主要擅长生成中、近景风格化人像。它们的主要功能是使基础模型可以在不使用画师串、较少的质量提示词的条件下获得相对稳定的图像质量，为提示词节约token空间。 本模型并非风格模型，在不同的提示词和生成条件下可能会有微妙的画风差异。 • Pony: People's Works (ppw) is a experimental fine-tuned model series, approximately 85% of the dataset comes from AI-generated images published by users on CivitAI. Since the earlier ppw dataset was built on images generated by Pony V6, the outputs of this series also carry some characteristics of Pony Diffusion. • This series uses standard Danbooru tags and is mainly optimized for generating stylized portraits at medium and close range. The primary effect of this model series is to allow the basemodel to achieve relatively stable image quality, without artist keywords or long quality tags, freeing up token space for prompts. These models are not style LoRAs. There may be subtle stylistic variations depending on different prompts and generating conditions. v7 v7版本对数据集结构做了较大幅度的调整，并且使用了不同的训练参数和训练策略，因此可能v7会不如原来的版本稳定。 The v7 version has undergone significant structural adjustments to the dataset, and utilizes different training parameters and strategies. As a result, v7 may be less stable than the previous versions. v-pred模型在civitAI在线生成器上的表现和吐司的在线生成表现完全不一样，相同参数完全无法复现。我也不知道是为什么...... The v-pred model's performance on the CivitAI online generator is completely different from online generation on TensorArt. The results are entirely unreproducible with a same parameters. I have no idea why... TensorArt version CivitAI ver. with same parameter on CivitAI with higher weight v7版本简介： 这是一个在前作的数据集基础上发展而来的图像质量LoCon，约90%-95%图片数据来自CivitAI上发布的图片。 它使模型可以在不使用画师串、使用较少的质量提示词的条件下获得相对稳定的图像质量，节约出更多的token空间，同时它还可以修复一部分模型固有的生成瑕疵。（但是不包括手部） 因为数据集选取的原因，生成的图片会带有Pony的质感。但因为它并不指向任何特定的画师、风格和绘画技法，所以在不同的提示词、模型条件下可能会有微妙的画风差异。 This is a generation quality LoCon developed based on the dataset from the previous work. About 90%-95% of the image data comes from CivitAI. It allows models to achieve relatively stable image quality without artist tags or using long quality prompts, freeing up more token space. Additionally, it can fix some inherent generation flaws of the model. (except for hands) Due to the dataset selection, the generated images exhibit a Pony-like style. However, since it does not reference any specific artist, style, or painting technique, there may be subtle stylistic variations depending on different prompts and checkpoint conditions. 数据集来源及许可证 | Dataset Source & License • 数据集中每一张图片都经过作者本人的人工筛选、分类和标注编辑，其中上千张图片经过人工的编辑、对细节瑕疵进行修正。 • 此模型为免费、开源模型，用户可以在私人设备上自行部署该模型。作者并不从模型出售中获取任何报酬。作者并不限制本系列模型用于商业生成服务或者生成图像用于商业用途，但是请注意配合使用的Checkpoint和其他LoRA的许可证限制。 • 请注意本模型的数据集由一个为约5000张AI图像构成的训练数据集和一个超过2000张图像构成的辅助数据集组成。其中，主数据集的图片大部分收集自AI社区。辅助数据集则包括公开的新闻图片、游戏CG和宣传图、3D渲染图和已购买的商业写真等。辅助数据集仅用于学习光影色彩、构图术语、人体解剖特征等一般通用知识，使用本模型不能还原辅助数据集中的版权内容。当前法律对这类数据的使用没有明确的统一规定，请有商用意向的本系列模型用户自行注意相关风险。 本数据集没有训练任何独立画师的数据，也没有标注任何画师ID信息（不排除AI错误标注的情况）。 • 另外，本模型不允许用作闭源商用、模型出售，也禁止用于闭源商用模型的融合。对于开源融合模型用于生成服务的情形不做限制，但是建议标注融合模型的出处。 • Every image in the dataset was manually reviewed, categorized, and annotated by the author. Among them, more than a thousand images were additionally hand-edited to manually correct fine-detail visual defects. • This model is free and open-source model, allowing users to deploy it on their personal devices. The author does not receive any compensation from selling the model. The author does not impose restrictions on using this model for commercial image generation services or generating images for commercial purposes. However, please be mindful of the license restrictions of the Checkpoint and other LoRAs used alongside this model. • Please note that this model’s dataset consists of a primary training set of approximately 5,000 AI-generated images and an auxiliary dataset of over 2,000 images. The majority of images in the main dataset were collected from AI communities. The auxiliary dataset includes publicly available news photographs, game CGs and promotional images, 3D-rendered images, and purchased photo sets. The auxiliary dataset is used solely to learn general-purpose knowledge such as lighting, composition terminology, and human anatomical features. This model cannot reproduce or restore copyrighted content from the auxiliary dataset. As current laws do not provide a clear, unified standard for the use of such data, users who intend to use this model for commercial purposes should be aware of and assess the associated legal risks on their own. This dataset does not include training data from any individual artist, nor does it contain explicit artist attributions (though AI mistagging cannot be entirely ruled out). • Additionally, this model is not permitted for use in closed-source commercial applications, model resales, or merged into closed-source commercial models. There are no restrictions on open-source merged models being used for image generation services, but it is recommended to credit the sources of any merged models.

+2

REMEMBER to follow the social networks and ask for your own model. New improved version for the Disney Studio model. This style aims to more faithfully imitate what we can see on the big screen in recent years. I've also noticed that generations significantly improve using "Cinematic" and "Cartoon". Hope you enjoy and have fun! Go to my Discord! --> https://discord.gg/zmnHnkCe Z-IMAGE TURBO 🔥 Trigger Word: "DisneyIZT" Add before the prompt-> "This image is a Disney-Pixar 3D animation style, featuring a stylized, cartoon character, high detailed, ultra 4k, masterpiece" for better results Illustrious XL Trigger Word: "DisneyStudioXL" (for Illustrious XL) and "Cartoon, cinematic" for better results. FLUX Trigger Word: "DisneyStudio" (for FLUX) and "Cartoon" for better results. PONY Use the activation word "DisneyStyleXLP" (For Pony) "Cartoon" for better results. ⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️ The best DISNEY PRINCESS COLLECTION here --> COLLECTION <--- All my STYLES here ---> STYLES <--- CHECK ALL MY MODELS here ---> ALL MY MODELS <--- ⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️ ADVICE!! If you want any COMISSION, contact me via: • Email: jeanc.romero.ib@gmail.com • Discord --> jeangcr • https://yeiyeiart.gumroad.com/ Check my comissions! ⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️ The best DISNEY PRINCESS COLLECTION here --> COLLECTION <--- All my STYLES here ---> STYLES <--- CHECK ALL MY MODELS here ---> ALL MY MODELS <--- ⬇️⬇️⬇️⬇️⬇️⬇️⬇️⬇️ ADVICE!! “If you want a commission of something in particular, whether it’s a style or a character on request, contact me via email: jeanc.romero.ib@gmail.com or write me by Discord --> jeangcr” Check my comissions!!:

+3

🥧 Pie Models – Pie Models is a collection of models, each named after a type of pie, offering a range of aesthetics from 2D anime to semi-realistic styles. These models are tailored for different visual outcomes and can be used individually depending on your desired style. • Apple Pie: Soft 2.5D Anime • Blueberry Pie: Semi-realistic [Tag Suggestion: realistic] • Cherry Pie: Anime–Western hybrid • Derby Pie: Semi-anime / Semi-Real • Elderberry Pie: Clean 2.5D Anime • Fudge Pie: Glossy 2.5D Anime • Grape Pie: Glossy Anime (Bright Stylized) • Honey Pie: Hybrid 2D / 2.5D Anime with semi-realistic elements • Impossible Pie: Semi-realistic [Tag Suggestion: realistic] – Update to Blueberry Pie • Jam Pie: Stylized 2D Anime (2D with some 2.5D ability) My Settings: • Positive Prompt: masterpiece, newest,absurdres,incredibly absurdres, best quality, amazing quality, very aesthetic • Negative Prompt: lowres,bad anatomy,worst quality,low quality,normal quality,bad hands,mutated,extra fingers,artifacts,disfigured, • Sampling Method: DPM++ 2M • Schedule Type: Karras • Steps: 30 • CFG Scale: 5 • Hi-Res Fix: Enabled for all images Use Adetailer if required.

+3

These checkpoints are for loras mainly, hence the style flexiblity, usually I test my loras on these, check if they look somewhat like the training material and I don't like to switch checkpoints around, so rather have an all in one checkpoint. The reason I upload them is to avoid ppl asking me how my stuff look like that. VAE download link: https://huggingface.co/stabilityai/sdxl-vae/resolve/main/sdxl_vae.safetensors?download=true V7 (no VAE): Exact same checkpoints used as V6, but different settings. This one is very sensitive to styles and prompts, reproduces faces more easily and more accurately. The reason I made this is because I've reached the limits of the pervious one and needed to upgrade, didn't think it would be this easy but!!! it still has the issue I made this in the first place, discoloration of the skin on flat color cartoon characters when trained strong enough, noticed this with some disney characters. Overall flexibility and accuracy improved, managed to mitigate the skin coloring to some degree, but the checkpoints that are good with realism will never be too good with the more flat cartoony look, sad. I don't use any quality tags with it, not in prompts or negatives. V6 (no VAE): It's a mix of SwedishFish v3 and Five Stars Illustrious v6 (i still have alexds9's permission to share merges of his models -I asked, just incase-). Flexible with styles and does better realistic characters. -I use Swedishfish because it has a secret ingridient and alex makes the other half of that secret sauce in his checkpoints- V5 had issues with face consistency, especially characters with realistic face proportions, at first I though it was the loras, later I found out it was actually the checkpoint since V6 can keep the faces really consistent without messing up the eyes or mouth (those 2 were the most common problems), and too often it just generated children with some realistic loras which was very annoying! It's still good with animated styles. V5 (no VAE): It's a mix of SwedishFish v3, Galena Blend v1.4 and Comics Fiction Engine Illustrious v1 (my man alexds9 gave me permission to share merges of his models). V4 had issues, like general lighting (too bright or too dark images) and discoloration with adetailer (I know thats a more common problem, but that occured more often with V4), This one has a pleasant base style while flexible and can handle realistic characters. Recommended quality tags: prompt: masterpiece, best quality, absurdres, very aesthetic, high quality, highres negatives: low quality, lowres, worst quality V4 (no VAE): Is the same as V3 except it's a mix of tOOnBabes and Illustrious 2.0 (50%-50%). It's been up for days now, but didn't publish it. Someone asked me very nicely to release it so made some showcase images. Recommended quality tags: prompt: masterpiece, best quality, absurdres, very aesthetic, high quality, highres negatives: low quality, lowres, worst quality V3: People were asking about this one for the style (especially with the Overwatch characters), so I'll share it. Wanted something flexible thats closer to original styles with loras without the forced anime to see if trained loras actually came out the way I wanted. It still does anime well, but atleast it won't force it on loras or when using style tags. Still not style neutral but good enough I guess. It's a mix between my tOOnBabes and Illustrious 1.0 (50%-50%). - I still don't know what I'm doing, just mixing stuff and see what comes out - No quality tags required! Atleast I don't use any in prompts or in negatives. V2 (Forgot to include VAE): Was playing around with some merges, so V2 has a different base style, is still flexible and anime is still there but not as strong. ps: I don't hate anime, there's just too much of it, not all of us are fans. psps: Maybe later i'll post some more showcase pics, don't think they're that important anyway. V1: Wanted to create a merge of illustrious that's not too anime heavy, I think I succseeded.

+1

Update: Please if you see any of my models somewhere else report or let me know! Thnx Trained on base model xl 1.0 + F1D + zit Please, check version info. "about this version" box on right. Feedback and ratings are much appreciated. Gun Models xl: Shotgun v1.0 Desert Eagle v1.0 Beretta 92FS v1.0 0.44 Magnum V1.0 AK-47 V1.0 M16 Series v1.0 Colt Revolver Gun v 1.0 STI 1911 series Uzi Handheld M134 Minigun Suppressor silencer Flamethrower MP5 M4/5 FN SCAR IWI Tavor FN P90 1921A Tommy Gun Gun Models F1D: Gun style F1D v1.0

+3

Cute Retro Girl is a checkpoint designed to generate adorable retro-style anime girls. It blends the charm of the 80s–90s with modern rendering: expressive eyes, soft highlights, clean lines, and rich color palettes. Perfect for creating cute characters with strong personality — from glossy cyber girls to flirty towel-clad cuties. The checkpoint excels at poses, gloss, neon glow, and clothing detail. Ideal for stylized scenes with a nostalgic yet futuristic twist. 💡 Additionally: The Cute Retro Girl model performs quite well even without using ADetailer, if you work with portraits, half-body, or medium-close shots. Faces appear clean, anatomically correct, and expressive. ADetailer is recommended when generating full-body characters in long-range compositions. For the face, it is necessary to use ADetailer 🎞️ Watch video example: https://civitai.com/images/70459219 📸 For full-length works or long-range plans: The examples use Img2img - Inpaint 🧪 Samplers: Recommended: • Sampler: Euler a / DPM++ 2M + Karras • Sampling Steps: 25–45 • CFG Scale: 5–7 (For portraits: 5–6)

+3

This LoRA is Studio Ghibli meets dark fairytales. Watercolor backgrounds and cel-shaded characters, ethereal lighting, fantastical natural environments, whimsical yet haunting. It was inspired by the timeless work of Hayao Miyazaki, Brian Froud, and Kay Nielsen. MidJourney V7: Used the best images from V1 + My personal MJ7 style to create the new dataset from scratch. The Aesthetic is preserved, and the versatility is greatly increased, but the colors are a little bit different as the Watercolor effects are more visible.

+3

A nude/NSFW capable model that focuses on images with a female primary subject. It is heavily biased towards nude images. Recommended Settings Sampler is dpmpp_2m (DPM++ 2M) and the scheduler is beta No VAE or CLIP is baked in. Use separate sources for those. Each sample/preview image contains the used workflow. Here's a quick article with simpler more beginner-friendly workflow. This is a recommended starting point. Regarding GGUF-versions If you had problems with GGUF-versions in Forge, it should now work. Please re-download the model to get an updated version. Early Access is enabled to support the development of new versions and models.

+3

Update: Please if you see any of my models somewhere else report or let me know! Thnx Trained on base model xl 1.0 + F1D + Illu + Pony + zit Works well with realistic stock photo model. Feedback and ratings are much appreciated. Feel free to check other similar models: 16mm (Super) Film Emulator style XL Anamorphic format film style XL (Hollywood Cinematic style) Cinematic text title (on screen) style XL Cinematic "color grading" motion picture film style XL Cinenaut XL (A TRUE CINEMATIC DIFFUSION) Cinematic Film Style XL Cinematic Photography Style XL Cinematic Kodak Motion Picture "Film Still" Style XL Cinematic Film Stock Footage Style XL Cinematic Vintage Film Style XL High Angle (from above) cinematic style XL Hyperrealism 3D Cinematic Style XL Kodak Film Grain (Cinematic) Photography style XL Infrared Cinematic Photography style XL Lens Flare Cinematic Reflection style XL Low Angle (from below) cinematic style XL Music Video cinematic style XL (Including 60's 70's 80's 90's, etc.) Point of view (pov + fps) cinematic subjective camera style XL Rotated View (Dutch Angle) Cinematic Style XL Realistic Photographic Film Stock Footage Style XL Set design style (cinematic scenic design) XL Split Diopter style (Deep Focus) XL Television advertisement cinematic Film style XL (including fashion models) Three-quarter view style XL Text message on screen bubble (cinematic style) XL Tilt shift photography style XL Unreal Engine style XL

+3