v2.0 2023.04.13<ul><li>I think it's easier to get a distinctive haircut or pink mesh in the hair.</li><li>I suggest you include hair-related keywords.</li></ul>Recommend options<ul><li>LoRA weight 0.6~0.8</li><li>Trigger words<ul><li>base<ul><li>Cure-Spicy</li></ul></li><li>Keywords for more detailed costumes:<ul><li>tied hair,multicolored hair,side ponytail,streaked hair,blue headwear,pantyhose,blue legwea,blue footwear,puffy sleeves,white gloves,navy blue footwear,brooch,jewelry,bangs,frills,flower,blue skirt,long skirt,high heels,jewelry,</li><li>Try different combinations of words and coordination!</li></ul></li></ul></li></ul>Use animefull based model<ul><li>Settings<ul><li><a target="_blank" rel="ugc" href="https://github.com/AUTOMATIC1111/stable-diffusion-webui">Stable Diffusion WebUI by Automatic1111</a></li><li>DPM++ Series(SDE Karras, 2M Karras, etc.)</li><li>about 20 steps, CFG scale 3.5~6.5</li><li>kl-f8-anime2.vae</li><li>CLIP skip = 2</li><li>Use highres. fix to get higher quality image<ul><li>Upscaler : Latent series</li><li>Denoising strength 0.50~0.65</li></ul></li></ul></li></ul>LoRA training info<ul><li>trained on <a target="_blank" rel="ugc" href="https://github.com/kohya-ss/sd-scripts">sd-scripts by kohya_ss</a>. Thank you a lot!</li><li>109 images of Cure-Spicy</li><li>replaced character feature tags with Cure-Spicy</li><li>resolution : 512x512</li><li>rank=alpha=16</li><li>batch size = 4, 7 epochs trained</li></ul>This LoRA may be compatible with animefull-based models (ex. AbyssOrangeMix Series...)This LoRA can generate nsfw images of Cure-SpicyThis LoRA can not generate nsfw images of huwakokokneThe above text is using a translatorキュアスパイシー（デリシャスパーティプリキュア）<ul><li>設定などは上記を参考</li><li>芙羽ここねは出力できません</li></ul>

Files with SHA256: c1fae287f63f0ef4d648a5875c7b3680fb408fc7e087f9b6966e463650507e17