<h3 id="important:-read-below-ldzgtjp7m"><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>Trigger word(s): ctianph1l, pink hair, yellow eyes, Clothing: white capelet, large breasts, underbust, blue skirt, frilled skirt, center frills, black thighhighs, red necktie, long sleeves, Battlesuit: white buruma, white sleeves, black gloves, black thigh boots, navel, groin, skin tight, covered shoulders,commissions: <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, ctianph1l, pink hair, yellow eyes, white capelet, large breasts, underbust, blue skirt, frilled skirt, center frills, black thighhighs, long sleeves, red necktie, outdoors, city, sunny, day, depth of field, looking at viewer, sitting, thighs, shiny skin, sweat, smell,</code></pre>Negative prompt ❌:<pre><code>sketch, 3d, pouch,</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<h1 id="faq:-cdomlyuo5">FAQ:</h1>Q: why not use character name as trigger word? A: I use unique token to prevent other things from being prompted when using triggerword Q: do you have other account in other websites? A: yes, tensor, pixai Q: what external generator do you use? A: a111 mostly, comfyui for img2vid

Philuffy Aingram フィルフィ・アインクラム PONY - Saijaku Muhai no Bahamut

the battle suit is undertrained, not enough data.

Philuffy_Aingram_pony_ct.safetensors

v1.0

<h3 id="important:-put-&quot;pouch&quot;-in-negative-prompts.-b5juclnpw"><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: Put "Pouch" in negative prompts.</h3>Trigger word(s): ctianph1l, pink hair, yellow eyes, Clothing: white capelet, large breasts, underbust, blue skirt, frilled skirt, center frills, black thighhighs, red necktie, long sleeves, Battlesuit: white buruma, white sleeves, black gloves, black thigh boots, navel, groin, skin tight, covered shoulders,commissions: <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, ctianph1l, pink hair, yellow eyes, white capelet, large breasts, underbust, blue skirt, frilled skirt, center frills, black thighhighs, long sleeves, red necktie, outdoors, city, sunny, day, depth of field, looking at viewer, sitting, thighs, shiny skin, sweat, smell,</code></pre>Negative prompt ❌:<pre><code>sketch, 3d, pouch,</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<h1 id="faq:-m7o7n4nsp">FAQ:</h1>Q: why not use character name as trigger word? A: I use unique token to prevent other things from being prompted when using triggerword Q: do you have other account in other websites? A: yes, tensor, pixai Q: what external generator do you use? A: a111 mostly, comfyui for img2vid

&lt;p&gt;the battle suit is undertrained, not enough data.&lt;/p&gt;

Files with SHA256: 3c38f050cdd25d4e167894c4e8cf90ef9b75059323114488f66a8c08627c9b4e