<h2 id="sarenda-the-moth-queen-(original-character)-ytticgoal">Sarenda, the Moth Queen (Original Character)</h2>It all started with a single image generated using the Pony model.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e84d8e9f-14f4-4e01-a175-4a7979d8e549/width=525/e84d8e9f-14f4-4e01-a175-4a7979d8e549.jpeg" /> (See <a rel="ugc" href="https://civarchive.com/articles/9297/how-to-train-character-lora-with-super-low-data-illustriouspony">this tutorial</a> for more details.)<h3 id="lora-versions:-gkjdkhkl4">LoRA Versions:</h3><ul><li>Illustrious v1 – Created using just one base image.</li><li>Pony v1 – Built from the original image plus 24 images generated with Illustrious v1. Trained over 10 epochs with 16 repeats per image.</li><li>Illustrious v2 – Same setup as Pony v1, but using the Illustrious model as the base.</li><li>Pony v2 – Same as Pony v1, but trained over 50 epochs with only 1 repeat per image.</li></ul>For generation tips, trigger words, and version-specific notes, please refer to the documentation provided with each version.<hr />I hope you enjoy using this LoRA! Feel free to rate it, leave comments, and share your creations — your feedback is always appreciated. Thank you!

Sarenda (Original Character)

How to generate correct image with WAI-NSFW-illustrious-SDXL generationUse at 0.8 strenght.Use a least "Srdtmq, hair over eyes" as trigger wordsRecommended settings for text-to-image:<ul><li>Sampler: Euler A (Karras)</li><li>Sampling steps: 20</li><li>CFG scale: 6</li></ul>Then, send your result to image-to-image with:<ul><li>Sampling steps: 30</li><li>CFG scale: 6</li><li>Denoising strength: up to 0.75 (lower values for a closer match to the original image)</li><li>Extras: ADetailer (for face, not for eyes cause she has hair over eyes).</li></ul>

Files with SHA256: f679326d952c66b4e643bbfa0aa479524dccfecaa4ef5b5ad4a65c0f65629747