<h1 id="ino-yamanaka-blank-period-(all-outfits)-v1">Ino Yamanaka - Blank Period (all outfits) v1</h1>Current: 04/30/2026 😈 illustrious Initial Release!Hey, Friend!! Long time no see. Took me a year to Finally get around releasing the official illustrious model haha. But its here... first stable build after the beta. I rebuilt the dataset with a focus on fidelity, prompt consistency and keeping the file lean. Synthetic samples were pruned and replaced with curated human artwork to reduce artifacts and improve full-body coherence. The main goal this time was matching the original anime/manga style rather than drifting into common fanart interpretations and I’m happy with how its turning out. Captions got a full cleanup so prompts behave more predictably across samplers and base models.- Streamlined outfits: Blank Period is now the default character presentation and no longer requires a separate trigger. Some variants have been removed to simplify tagging and focus . watch out for my official workflows with Nvidia and seedvr2 upscaler options!- Removed Tags: Blank period, Naruto Shippuden, Florida Outfit and their related outfit tags as well as long skirt, long hair and cropped.- New wardrobe tags: black underskirt and fishnet legwear.- Default earrings: Earrings are now baked into the base look. Add earrings to your negative prompt to remove them.- Workflow: I’m uploading my custom ComfyUI workflow as well. The sample images in this release were upscaled using the NVIDIA Super Resolution node and given slight film grain for a retro vibe—soft, natural lines instead of the usual razor-sharp AI perfection. I still plan to test with Forge and ADetailer later to see how much further it can go.<h2 id="story">Story</h2>Hey, welcome friend! Beyond excited to share my second lora attempt where I switch up my training process hopefully showing improvements from my Chocho model. Biggest difference was using <a target="_blank" rel="ugc" href="https://civarchive.com/models/260267">Animagine XL V3.1</a> as the base model instead of <a target="_blank" rel="ugc" href="https://civarchive.com/models/269232">AAM XL (Anime Mix)</a>. Ino. In my humble opinion. Is best girl in Naruto. So I took special care in making this lora as I use it for personal use. I've used other loras for months now for amazing results but it doesn't seem like anyone made a version of her during the blank period (Which I see as peak design for the majority of the cast). You can prompt for all outfits she wore during this period, her Naruto Shippuden outfit, plus a bonus outfit I made myself. All outfits can be worn with a pony tail or hair down long. I decided to make an SDXL version of the lora but I do plan to release a SD1.5 version as well.04/20/24 💥SD1.5 Update!!: Added SD1.5 version!!!! I personally don't think its as good as the SDXL model but its more lightweight and seems to get the job done for the most part. Same triggers as the SDXL version though I haven't done really any testing... Let me know how it its!10/31/24 🐎 Pony Update: Happy Halloween have Fun! 👻11/4/2024 🦄 Pony Update: Finally got around to posting the Pony Lora for this character. Been sitting on it for a while but it worked pretty good after some months of testing. I still consider this unfinished however, so this is just a release candidate. Let me know how you like it!12/25/2024 Pony Update: 🎠Released Finished version of the pony lora! I would say, in my humble opinion, this is the best model yet. This model overall produces the best quality images when it comes to anatomy, prompt adherence and detail. Added extra trigger word for the wedding outfit, but the rest should be the same. Post. your. pics!!! Feedback is welcome. Happy holidays 🎅 Follow me on Bluesky @burnthevillage.bsky.social: <a target="_blank" rel="ugc" href="https://bsky.app/profile/burnthevillage.bsky.social">https://bsky.app/profile/burnthevillage.bsky.social</a>04/30/2025 illustrious (Beta) Drop: Not sure all the illustrious hype, but I've observed Good prompt adherence. Tested with WAI0731's model. The model doesn't seem to be able to handle hands as well as pony. Let me know how it is. 😇Key Words:<ul><li>ponytail: ponytail, flower hair,</li><li>No ponytail: long hair,</li><li>Blank Period: blank period, long skirt, purple apron skirt, purple high-collared blouse, mesh arm warmers, midriff, navel, toeless footwear,</li><li>Naruto Shippuden (Note: the spelling is little different): naruto shippuuden , alternate costume, purple high-collared blouse, midriff, navel, miniskirt, bandages, thigh holster, thigh strap, sandals, mesh arm warmers,</li><li>Chef outfit: chef uniform, mesh arm warmers, white apron, toeless footwear,</li><li>Wedding Dress: pink shirt, emerald necklace, purple bracelet, long skirt, purple apron skirt, shirt tucked in,</li><li>Florida Outfit ;) : florida outfit, torn shorts, denim shorts, purple high-collared blouse, mesh arm warmers, toeless footwear, midriff, navel,</li><li>Framing the shot: full body, upper body, close-up, cropped</li><li>Wedding Dress (Pony): wedding outfit, pink shirt, emerald necklace, purple bracelet, long skirt, purple apron skirt, shirt tucked in,</li></ul>Suggested settings (SDXL):<ul><li>I wanted the pictures to come off as natural as possible, so I didn't use certain tools like Controlnet, face restore or Inpainting but I used ADetailer for some examples.</li><li>Outputs are typically just what you would get from putting in a prompt and using <a target="_blank" rel="ugc" href="https://civarchive.com/models/178518/eyeful-or-robust-eye-detection-for-adetailer-comfyui">ADetailer</a> for the eyes only in combination with a high-res fix, which I suggest a <a target="_blank" rel="ugc" href="https://openmodeldb.info/models/4x-realesrgan-x4plus-anime-6b">4x upscaler.</a></li><li>I trained the LORA with Clip Skip 2.</li></ul>I tested with a few different XL models including a lightning model which I included best results (However the testing was brief for each. I swear I / you can make better images :D). Decided to dust of Please feel free to leave any feedback!Suggested settings (Pony):I liked to use some Negative Embeddings when prompting:<ul><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/119032?modelVersionId=641519">unaestheticXL | Negative TI</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/332646/pony-pdxl-negative-embeddings?modelVersionId=720175">Pony PDXL Negative Embeddings</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/models/118418/negativexl">negativeXL</a></li></ul>When using <a target="_blank" rel="ugc" href="https://civarchive.com/models/448716/boleromixpony?modelVersionId=1161495">boleromix</a>:<ul><li>I like to play around with CFG of 2 - 3.5</li><li>Sampling steps of 25 - 30</li><li>Sampler euler-a Turbo</li><li>4x Upscaler (of your choice. I like <a target="_blank" rel="ugc" href="https://openmodeldb.info/models/4x-realesrgan-x4plus-anime-6b">RealESRGAN_x4Plus Anime 6B</a>)</li></ul>Suggested settings (illustrious Beta):When using <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-nsfw-illustrious-sdxl">wai-nsfw-illustrious-sdxl v13</a>:<ul><li>I just stuck with CFG of 7</li><li>Sampling steps of 30</li><li>Sampler euler-a Turbo</li><li>4x Upscaler (Pick one you like. I like <a target="_blank" rel="ugc" href="https://openmodeldb.info/models/4x-realesrgan-x4plus-anime-6b">RealESRGAN_x4Plus Anime 6B</a>)</li><li>I started adding "eye iris" to the negative prompt for that added level of detail</li></ul>Suggested settings (illustrious):<a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-illustrious-sdxl?sync-account=green">wai-nsfw-illustrious-sdxl v17</a><a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-illustrious-sdxl?sync-account=green">:</a><ul><li>I like CFG of 6 - 7</li><li>Sampling steps of 28 - 30</li><li>Sampler DPMPP_2M or euler a (try both)</li><li>I used a upscaler new to me but requires nvidia GPU <a rel="ugc" href="https://github.com/Comfy-Org/Nvidia_RTX_Nodes_ComfyUI.git">RTX Upscaler Node</a></li><li> After being spoiled with more powerful models I got used to perfectly drawn faces, but it seem sdxl, pony, ill all have issues with face. I recommend upscalers, face and eye detailer</li><li>All Clothing is modular, so mix and match however you want.</li><li>prompting is pretty much the same except mentioned removed tags</li></ul>Have Fun,Florida-Man<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/10565e32-19e6-417b-9e2c-062c12124d72/width=525/10565e32-19e6-417b-9e2c-062c12124d72.jpeg" /><a target="_blank" rel="ugc" href="https://ko-fi.com/904floridaman"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/eb2b6b46-efc7-482b-9476-132d04e8eb84/width=525/eb2b6b46-efc7-482b-9476-132d04e8eb84.jpeg" /></a>Your contribution is greatly appreciated and helps me to create more content. If you like my model, please consider donating: <a target="_blank" rel="ugc" href="https://ko-fi.com/904floridaman">https://ko-fi.com/904floridaman</a>I'm open to commissions.

Ino Yamanaka - Blank Period [all outfits] (山中いの, Yamanaka Ino)

<ul><li>RC4 — Initial Release</li><li>First public release following the beta phase. </li><li>Training dataset overhauled: synthetic samples pruned and replaced with curated, high-quality human artwork to improve fidelity and reduce artifacts. </li><li>Caption tags cleaned and standardized for more consistent prompting behavior. </li><li>Hyperparameters tuned to maximize detail retention without increasing file size or bloat.</li><li>Outfit tags streamlined: Blank Period is now the default character presentation. Naruto Shippuden and Florida Outfit variants have been removed to simplify tagging and improve output consistency. :'(</li></ul>

Files with SHA256: 19035c1d589fb79a27229832a08f2d878c0af033ff90431de4ba1733ad5935e6