<h2 id="pixie-queen-from-magic-the-gathering-(legends)-r4iikoyjt">Pixie Queen from Magic the Gathering (Legends)</h2>What would happen if I used a single low-resolution, not very detailed card illustration to train a LoRA?<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0f6f05a1-78b7-4b10-bfe3-9b7a8113b02e/width=525/0f6f05a1-78b7-4b10-bfe3-9b7a8113b02e.jpeg" />It’s not perfect — I think I can do better, even with just one image. It’s definitely experimental, but I’m happy with it and wanted to share!Usage Instructions Use with the following settings:<ul><li>Basic trigger words: <code>pqmtg, green hair, fairy wings, green lips</code></li><li>Optional for eyes: <code>no pupils, no sclera</code></li><li>Clothing trigger words: <code>choker, errings, jewelry, staff</code></li><li>Strength: 0.8–0.9</li><li>Add skin color (pale skin for exemple) cause the Lora want her to have green skin.</li><li>Add Solo to the prompt</li></ul>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 !

Pixie Queen (Magic The Gathering - Legends)

How to generate correct image with Arthemy Comics IllustriousRecommended settings for text-to-image:<ul><li>Sampler: Euler A (Karras)</li><li>Sampling steps: 30</li><li>CFG scale: 8</li></ul>Then, send your result to image-to-image with:<ul><li>Sampling steps: 40</li><li>CFG scale: 8</li><li>Denoising strength: up to 0.75 (lower values for a closer match to the original image)</li><li>Extras: ADetailer (if you want her to have the original eyes, dont use Adetailer for eyes)</li></ul>

Files with SHA256: 969ebc29a2963242b94cadecb1780168c28429ad6e4b933848276c280700a7ef