<h3 id="important:-read-below-ctj6jia8u"><a target="_blank" rel="ugc" href="https://ko-fi.com/ctian"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/cf614846-63c8-4596-a5fd-f5962d69ae75/width=525/cf614846-63c8-4596-a5fd-f5962d69ae75.jpeg" /></a>IMPORTANT: READ BELOW</h3>You can prompt her darker skin colour with prompt "tan"Trigger word(s): ctianfranc3sca, Clothing: white uniform, white ribbon, neck ribbon, striped panties, Cat: cat ears, cat tail,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, ctianfranc3sca, (tan:0.3), white uniform, white ribbon, neck ribbon, facing viewer, looking at viewer, outdoors, sunny, day, city, depth of field, striped panties,</code></pre>Negative prompt ❌:<pre><code>sketch, 3d, claws</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

Francesca Lucchini フランチェスカ・ルッキーニ PONY - Strike Witches ストライクウィッチーズ

17 additional training images added to training datasettraining images include cat girl Francescaquality improved please use this version because cat girl

Francesca_Lucchini_updated_v2_ct.safetensors

v2.0 UPDATED

v1.0

<h3 id="francesca-lucchini-pony-strike-witches-t4ehyj358">Francesca Lucchini フランチェスカ・ルッキーニ PONY - Strike Witches ストライクウィッチーズ</h3>kofi commission AkumakushuTrigger word(s): ctianfranc3sca, Clothing: white uniform, white ribbon, neck ribbon, striped panties, Cat: cat ears, cat tail,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, ctianfranc3sca, (tan:0.5), white uniform, white ribbon, neck ribbon, looking at viewer, outdoors, sunny, day, city, depth of field, cat ears, cat tail, claw pose, chestnut mouth, upper body,</code></pre>Negative prompt ❌:<pre><code>sketch, 3d, claws</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;17 additional training images added to training dataset&lt;/p&gt;&lt;p&gt;training images include cat girl Francesca&lt;/p&gt;&lt;p&gt;quality improved &lt;/p&gt;&lt;p&gt;please use this version because cat girl&lt;/p&gt;

Files with SHA256: 2f7c474f4ad19747da8ab70b389dfd42241bec510af4ab03b9ed3b97189f927a