✨ Hanasaka Ibuki from Kotowarenai Kaichou wa Tomoe-kun ni Dake shite Agetai
— She is one of the more playful and energetic girls from Kotowarenai Kaicho wa Tomoe-kun ni Dake Shite Agetai. She is highly expressive and often carries a mischievous, teasing smile, giving her the kind of presence that makes it feel like she is always about to mess with someone. At the same time, she can become surprisingly flustered herself, which makes that confident playful side even more fun to watch. Her lively expressions and distinctive ribbon-heavy hairstyle also make her stand out quite easily among the cast.
⚠️Turn off your R/X/XXX Filter to see hidden JUICY Loras in >HERE<
🖼️ How I Generate My Images
I made a full article that explains how I generate my images — including my settings, prompting tips, and even my potato PC setup 😅
👉 Check it out here if you’re curious or want to try the same method.
🔞 NSFW gallery (18+)
Check my account here to see the spicy showcase that I can't show here, if any.
🎨 Prompt Tips
These are optional starting points for handling a few behaviors of this version. You may still need some retries depending on the checkpoint, prompt, and LoRA strength.
🖤 Manga-like result
For a result closer to the original doujin’s greyscale manga appearance, avoid positive tags that strongly push anime coloring, vivid colors, or an anime screenshot style. It is also better to avoid character or outfit color tags, since they may encourage the checkpoint to colorize the image.
Recommended:
✅ Positive prompt: greyscale
🌈 Controlling appearance colors
Since the dataset is entirely greyscale, this LoRA does not contain reliable color information for her hair, eyes, or other appearance details.
If you want specific colors, directly include the corresponding color tags in your positive prompt.
The colors used in my showcase are simply my own interpretation, so feel free to change them depending on your preference.
🎀 Controlling the hair ornament color
When changing the color of her hair ornament, some checkpoints may interpret a tag such as red hair ornament as instructions for both a red ornament and red hair.
For example, if you want a red hair ornament while keeping her hair color unchanged, try:
✅ Positive prompt: red hair ornament
🚫 Negative prompt: red hair
You can apply the same approach with other colors if the checkpoint starts affecting her actual hair instead of only the ornament.
🎓 Suppressing extra blazer decorations
Some checkpoints tend to associate blazers with school-uniform decorations and may add extra details such as an emblem, badge, or logo even when those details are not part of the intended outfit.
To reduce that behavior, try:
🚫 Negative prompt: school emblem, emblem, badge, logo
This is mainly a checkpoint bias rather than an outfit detail leaking from the LoRA. It is especially useful when the generated blazer keeps gaining unnecessary symbols or decorations on its own.
⚖️ LoRA strength and outfit flexibility
A weight of 1 can still work with a very different outfit, especially when the new outfit is clearly different from her default one.
Only lower the LoRA weight gradually when outfit details keep leaking even after using clear outfit tags and negative prompts. Lowering it too much may also weaken her face, hair, and overall character identity.
📊 Lora Project
If you want to know what characters that I'm covering next you can check these link :
👉 Personal Project
👉 Commision Project
If this model sparked something in you, show some love with a like, review, or buzz.
Now go—
Be wild
Description
⚫ Greyscale-only dataset
This LoRA was trained using greyscale manga images only. There are no colored images in the dataset.
Because of that, the appearance colors shown in my showcase are based on my own taste rather than strict source accuracy. Hair color, eye color, hair ornament color, and similar appearance details should therefore be treated as my interpretation rather than colors learned directly from the source.🎓 Default uniform colors follow Chitose
The default outfit is the main exception to the personal color choices used in my showcase.
Since Ibuki and Chitose are from the same series and wear the same school uniform, I use the same color scheme that I used for Chitose's default uniform. This keeps their shared uniform visually consistent instead of giving the same outfit two unrelated color schemes.
The dataset itself is still greyscale-only, so these colors are not actually trained into the LoRA and can still be changed through prompting.💇 Two hairstyle variations
The dataset contains two main hairstyle variations:
- short twintails
- medium hair.
Both styles are part of the original dataset coverage, so you can switch between them depending on which version of Ibuki you want to generate. 🎀 Hair ornament color is flexible
Because the dataset does not contain color information, the hair ornament is not tied to one trained color.
You can assign its color through prompting depending on the look you want. However, some checkpoints may misunderstand combinations such as red hair ornament and also change the character's hair color, so additional negative prompting may be useful when that happens.🎓 Possible blazer decoration bias
Her default uniform includes a blazer, and some checkpoints tend to associate blazers with additional school-uniform details such as logos, badges, or emblems.
These extra decorations are not intended to be defining parts of her outfit. Their appearance can depend heavily on the checkpoint being used, and they can usually be reduced through negative prompting.👗 Limited skirt coverage
A large portion of the dataset consists of upper body images, which means the lower half of her default uniform has less visual coverage than her face, hair, and upper-body clothing.
Because of that, the skirt design may not stay perfectly consistent between generations. Small differences in the plaid pattern, folds, shape, or other skirt details can appear depending on the checkpoint and prompt.
The general school-uniform look should still remain recognizable, but the skirt is one of the less stable parts of this version.As usual, the result can still depend on your checkpoint, prompt, LoRA strength, CFG scale, and generation settings.