<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7866ed20-3126-40b0-8f76-a783f716943b/original=true/7866ed20-3126-40b0-8f76-a783f716943b.jpeg" /><h2 id="akamine-satsuki-from-fuuki-iinchou-wa-mochikaerarete-moteasobarete">✨ Akamine Satsuki from Fuuki Iinchou wa Mochikaerarete Moteasobarete</h2>— She is the strict discipline committee type who looks like she can silence a whole classroom with one glare. She has that serious, no-nonsense aura at first glance, but the fun part is how easily that composed image can crack once the situation stops going her way. Her sharp expression, confident posture, and unexpectedly cute reactions make her feel both intimidating and easy to tease. A nice pick if you like stern schoolgirl characters with a softer, more embarrassed side hiding underneath.<hr />483797 &amp; 529931 &amp; 627190If you know, you know~<h1 id="greaterdoujinshi-collectionless"><a target="_blank" rel="ugc" href="https://civarchive.com/collections/11862568">&gt;Doujinshi COLLECTION&lt;</a></h1><h1 id="turn-off-your-rxxxx-filter-to-see-hidden-juicy-loras-in-greaterhereless">⚠️Turn off your R/X/XXX Filter to see hidden JUICY Loras in <a target="_blank" rel="ugc" href="https://civarchive.com/collections/13797500">&gt;HERE&lt;</a></h1><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e4c11a1c-1328-4124-9572-eee4f03663f0/original=true/e4c11a1c-1328-4124-9572-eee4f03663f0.jpeg" /><h2 id="how-i-generate-my-images">🖼️ How I Generate My Images</h2>I made a full article that explains how I generate my images — including my settings, prompting tips, and even my potato PC setup 😅👉 <a target="_blank" rel="ugc" href="https://civarchive.com/articles/17562">Check it out here</a> if you’re curious or want to try the same method.<h2 id="nsfw-gallery-(18+)">🔞 NSFW gallery (18+)</h2>Check my account <a target="_blank" rel="ugc" href="https://sociabuzz.com/zetsubou">here </a>to see the spicy showcase that I can't show here, if any.<h2 id="prompt-tips">🎨 Prompt Tips</h2>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.<h3 id="manga-like-result">🖤 Manga-like result</h3>For a more manga-like result, avoid tags that strongly push anime coloring or an anime screenshot look in the positive prompt, also avoid any color tags. You also need to use : ✅ Positive prompt: <code>greyscale</code><h3 id="hair-color">🎀 Hair Color</h3>To make her hair color closer to canon, I recommend using these tags : 🚫 Negative prompt: <code>(multicolored_hair:1.2),(colored_inner_hair:1.2)</code><h3 id="mole-details">🔎 Mole details</h3>If you want the mole placement to be more accurate, you need to keep the hires fix denoising strength very low. In my experience, it works best below <code>0.1</code>, or even at <code>0</code> if you only want to upscale without adding denoising.If you care about getting her mole details right, try these settings: ✅ Positive prompt: <code>mole_under_eye, mole_on_breast, mole_on_thigh</code> 🚫 Negative prompt: <code>(multiple_moles:1.5)</code> ⚙️ CFG: <code>6</code> 💪 Denoising strength (hires fix): <code>0.05</code>If you do not care too much about inaccurate mole placement, you can just use my usual denoising strength of <code>0.3</code> instead.<h3 id="shirt-+-skirt-tucked-in-bias">👕 Shirt + skirt / tucked-in bias</h3>This version may sometimes tuck the top into the skirt, especially when prompting shirt + skirt combinations. This happens because the original source only shows that specific outfit combo with the tucked-in detail, even though the dataset also includes other outfits.If you want to avoid it, you can try using <code>untucked_[clothing type]</code>, such as <code>untucked_sweater</code>.For an untucked sweater look, try: ✅ Positive prompt: <code>untucked_sweater</code> 🚫 Negative prompt: <code>(hem_peeking_out:1.3), (untucked_shirt:1.3), sweater tucked in</code>This should help, but it may still need a retry sometimes because the tucked-in look is a dataset bias.<h3 id="default-outfit-details-appearing-on-other-outfits">🧥 Default outfit details appearing on other outfits</h3>When a default outfit detail keeps appearing in a different outfit, add the exact leaking detail to the negative prompt.For example, if the pleated details on her skirt appearing when you do not want it: 🚫 Negative prompt: <code>pleated_skirt</code>Use the specific leaking tag instead of adding many unrelated negative tags.<h3 id="lora-strength-and-outfit-flexibility">⚖️ LoRA strength and outfit flexibility</h3>A weight of <code>1</code> 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.<h2 id="lora-project">📊 Lora Project</h2>If you want to know what characters that I'm covering next you can check these link : 👉 <a target="_blank" rel="ugc" href="https://docs.google.com/spreadsheets/d/1oM6Xk0k5sBCw2CQfIFMzQQuuC6lHdOOnw3zES9phBm8/edit?gid=1864372839#gid=1864372839">Personal Project</a> 👉 <a target="_blank" rel="ugc" href="https://docs.google.com/spreadsheets/d/1k15ya9i0TjAHJnBpVfPSsjHsngpjkLsAZ11IcOT0hc0/edit?gid=0#gid=0">Commision Project</a>If this model sparked something in you, show some love with a like, review, or buzz. Now go— Be wild

Akamine Satsuki | Fuuki Iinchou wa Mochikaerarete Moteasobarete/風紀委員長は持ち帰られて弄ばれて [chinetsu]

🎨 Mostly greyscale source<pre><code>Most of the dataset is based on greyscale source images, so some color details may need a bit of prompt guidance. I used the available cover images as color references, but those covers have unusual purple lighting/effects, so her hair may sometimes generate with a slightly two-tone look when prompted normally.</code></pre>🔎 Small detail sensitivity<pre><code>Details like her moles can be tricky, especially when using higher hires fix denoising. If you want the placement to stay closer to the source, you may need to use lower denoising strength and a more specific prompt setup.</code></pre>👕 Shirt + skirt bias<pre><code>There is also a small tucked-in top bias. In the original source, the shirt + skirt combo is always shown with the top tucked into the skirt, so similar prompts may sometimes inherit that detail too.</code></pre>These are all still controllable with prompting, but depending on your checkpoint, prompt, LoRA strength, and hires settings, you may need a few retries. I added more specific prompt tips below for handling these cases.

Files with SHA256: e4c2aa0badbb5e3b6a0ff71dbf793f215dec9a1bc34e537b72077436dce882d5