Trained exclusively using my own screenshots taken by the IN-GAME CAMERA.Update:<ul><li>Dataset 2.0: Added ~80 unique images (360 total) to help seperating the character from the style more, as well as adding more NPC outfits (~20 total, all trained pretty well) and scenery to increase model capability and background variety.</li></ul><ul><li>Dataset 1.1: Added more images for the two girls (first two images in 1.1 (Anima) showcase) who was not well trained in 1.0 dataset.</li><li>Added V1.0 (Lite): Only keep necessary blocks for the style. This fixes detail loss (especially for hair style and clothing) and body proportions for some characters.</li></ul>Recommended Settings:V1.1&amp;2.0 (Anima): Trained on <a target="_blank" rel="ugc" href="https://civarchive.com/models/2458426/anima?modelVersionId=2945208">Anima base-v1.0</a>Sampler: ER SDE or Euler*<span style="background-color:rgb(26, 27, 30);color:rgb(141, 141, 143);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Scheduler: Beta or Simple*Steps: 30 or 12*CFG: 4-5 or 1*LoRA Strength: 0.8-1.0(*) with Turbo LoRA~Triggers:Character:<pre><code>npc girl, zenless zone zero</code></pre><pre><code>npc girl \(zenless zone zero\)</code></pre>or natural language saying something like<pre><code>A npc girl from zenless zone zero</code></pre>Style (Use natural language for 2.0):<pre><code>zenless zone zero \(style\), 3d</code></pre>or natural language saying something like<pre><code>A zenless zone zero style 3d render</code></pre>~Extra Tip: since Turbo LoRA with CFG = 1 can't take any negative prompt, use<pre><code>no hair ornament, no hairclip, no hair bow</code></pre>to remove unwanted hair accessories or just use a different seed.~V1.0: Trained on <a target="_blank" rel="ugc" href="https://civarchive.com/models/1232765?modelVersionId=1389133">Illustrious XL 1.0</a>Sampler: Euler a<span style="background-color:rgb(26, 27, 30);color:rgb(141, 141, 143);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Scheduler: NormalSteps: 28-30CFG: 6-7LoRA Strength: 0.9-1.0~Trigger words:Character Tag:<pre><code>zzznpc girl</code></pre>Style Tag:<pre><code>3d</code></pre>~Random Character Generator (Outdated, but still works):If you have Dynamic Prompts, or any kind of wildcard nodes/plugins, here is an one liner to generate a random character:<pre><code>{closed mouth|parted lips}, {jitome|tareme|tsurime}, {brown|grey|yellow|pink|blue|red|aqua} eyes, {upper body|cowboy shot, {black|red|brown|pink|blue} skirt, pleated skirt, {plaid skirt|}}, {{blue jacket, open jacket|aran sweater, red sweater, v-neck|yellow cardigan, v-neck|}, red bowtie, {plaid bowtie|striped bowtie|}, white shirt, collared shirt, school uniform|{black|grey|blue} hoodie, hood down, {black|grey} hood, drawstring, clothes writing, {{black|grey} elbow patches|}, casual}, {brown|blonde|green|orange|blue|black|purple|grey|aqua} hair, {long hair, {twintails|ponytail|curly hair}|short hair, {bob cut|flipped sidelocks|}|medium hair, {wavy hair|flipped hair|low twin braids}}, {{short|swept|blunt|double-parted|arched} bangs|bangs pinned back|middle part, parted hair, forehead}, {medium breasts|small breasts|flat chest}, looking {ahead|to the side|at viewer}, {from side|from above|from below|}, {indoors|outdoors, {day|night|dawn|dusk|}}, {sidelighting|backlighting|}</code></pre>Note: Because some outfits are inconsistant and barely trained, I didn't put them in here. The main focus of the LoRA is the style.~Time of Day (light color, only for outdoors):<pre><code>dawn | day | dusk | night</code></pre>Light Direction (affect shading on character's face, can also put both in negative prompt to get front lighting):<pre><code>sidelighting | backlighting</code></pre>Shadow (more likely to put character under direct sunlight):<pre><code>shadow</code></pre>Shade (around character or in the shade):<pre><code>shade</code></pre>Reflections (these are introduced for training after V1.1):<pre><code>reflective floor, reflective liquid, reflective surface, reflection</code></pre>Some useful negatives that may improve the results:<pre><code>low poly, model clipping, bad perspective, aliasing</code></pre>

Zenless Zone Zero NPC Girl Style

Added ~80 more images to the 1.1 dataset (360 unique images in total), while also changing the training strategy. Also added few more outfits (~20 total unique outfits).

Files with SHA256: 108202e3ce764e509a908e20dadd6f11754a1b73da1a34cde2245ca2e3d5f87a