<h3 id="important:-read-below-cwo14yz4w"><a target="_blank" rel="ugc" href="https://ko-fi.com/ctian"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/eeabc2aa-b4d7-4c92-9cf5-85ae14d96764/width=525/eeabc2aa-b4d7-4c92-9cf5-85ae14d96764.jpeg" /></a>IMPORTANT: READ BELOW</h3>Trigger word(s): ctianm0rin0, Clothing: red serafuku, white sailor collar, ribbon, black skirt, long sleeves, pleated skirt,commission: <a target="_blank" rel="ugc" href="https://ko-fi.com/ctian">https://ko-fi.com/ctian</a> 😼 Example Positive prompt ✅:<pre><code>score_9, score_8_up, score_7_up, source_anime, masterpiece, 1girl, ctianm0rin0, red serafuku, white sailor collar, ribbon, black skirt, pleated skirt, long sleeves, looking at viewer, city, outdoors, depth of field, leaning forward, chestnut mouth,</code></pre>Negative prompt ❌:<pre><code></code></pre>Recommended resolution, aspect ratio for sdxl, pony 📷768 x 1344: Vertical (9:16)832 x 1216: Portrait (4:5) **1024 x 1024: Square (1:1)1182 x 886: Photo (4:3)1254 x 836: Landscape (3:2)1365 x 768: Widescreen (16:9)Work flow, procedure 🔧1. Collect dataset (booru, anime screencaps, search)2. Clean up dataset using upscalers, inpainting and photoshop3. Tag dataset using booru <a target="_blank" rel="ugc" href="https://github.com/starik222/BooruDatasetTagManager">github</a>4. Train LoRa in kohya ss locally <a target="_blank" rel="ugc" href="https://github.com/bmaltais/kohya_ss">github</a>5. Test epochs using xyz grid in controlnet to ensure LoRa is not overfitted/underfitted.6. Submit model to CivitAI <a target="_blank" rel="ugc" href="https://civarchive.com/user/ctian">profile</a>do not ask me training parameters, check lora metadata for training parameters

Seria Morino 守野 せりあ PONY - Gekidol ゲキドル

all training data from screencaps. style will be biased towards anime style

Seria_morino_pony_ct.safetensors

v1.0

<h3 id="seria-morino-pony-gekidol-g94xlryna">Seria Morino 守野 せりあ PONY - Gekidol ゲキドル</h3>@LuckyDonkeyTrigger word(s): ctianm0rin0, Clothing: red serafuku, white sailor collar, ribbon, black skirt, long sleeves, pleated skirt,commission: <a target="_blank" rel="ugc" href="https://ko-fi.com/ctian">https://ko-fi.com/ctian</a> 😼 Example Positive prompt ✅:<pre><code>score_9, score_8_up, score_7_up, source_anime, masterpiece, 1girl, ctianm0rin0, red serafuku, white sailor collar, ribbon, black skirt, pleated skirt, long sleeves, looking at viewer, city, outdoors, depth of field, leaning forward, chestnut mouth,</code></pre>Negative prompt ❌:<pre><code></code></pre>Recommended resolution, aspect ratio for sdxl, pony 📷768 x 1344: Vertical (9:16)832 x 1216: Portrait (4:5) **1024 x 1024: Square (1:1)1182 x 886: Photo (4:3)1254 x 836: Landscape (3:2)1365 x 768: Widescreen (16:9)Work flow, procedure 🔧1. Collect dataset (booru, anime screencaps, search)2. Clean up dataset using upscalers, inpainting and photoshop3. Tag dataset using booru <a target="_blank" rel="ugc" href="https://github.com/starik222/BooruDatasetTagManager">github</a>4. Train LoRa in kohya ss locally <a target="_blank" rel="ugc" href="https://github.com/bmaltais/kohya_ss">github</a>5. Test epochs using xyz grid in controlnet to ensure LoRa is not overfitted/underfitted.6. Submit model to CivitAI <a target="_blank" rel="ugc" href="https://civitai.com/user/ctian">profile</a>do not ask me training parameters, check lora metadata for training parameters

&lt;p&gt;all training data from screencaps. style will be biased towards anime style&lt;/p&gt;

Files with SHA256: 42a3e350da017d3e67fa6a006ae1408ec5acc3bee219c37101e62edb950b6a31