<a target="_blank" rel="ugc" href="https://www.buymeacoffee.com/pettankoutei"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/70cb8ffb-433c-44f0-98b7-43cf22ffb220/width=525/70cb8ffb-433c-44f0-98b7-43cf22ffb220.jpeg" /></a>If you like my content consider supporting me! 😊1 Cookie = Early access to future releases! 🍪❤️Lycoris trained on the style of Wagashi for Pony Diffusion v6 XL based models.Requested by a supporter! 🍪❤️<h3 id="heading-569"></h3><h3 id="heading-45">Pony version</h3>There is no trigger word, but you may use <code>aisaka taiga</code> to generate the character in the artist style. Suggested weights are between 0.8 and 1.0, but I personally used 0.9 in all my samples.<h3 id="heading-18">SD 1.5 version</h3>There is no trigger words and can be used at weight 1.0.The NAI version of this model also captures the style very well, but the default coloring style is much more influenced by the checkpoint and quality tags. Testing on the AnyLora checkpoint (since it is very close to base NAI), my default prompts <code>(masterpiece, best quality:1.2)</code>gives the images a smoother coloring typical of SD1.5 anime models. To get closer to the style of Wagashi, i used tags such as <code>traditional media, sketch</code> with weights between 1 and 1.3 (too high and backgrounds may be affected).<h2 id="heading-1488">Disclamer</h2>All the models that I have developed and published on this platform are for entertainment purposes only. They are derived from the base Stable Diffusion model, which means it also includes its inherent issues, limitations and unfiltered capabilities. With that in mind, I am in no way responsible for any content that the user creates with malicious intent while using any of these models. The user assumes complete responsibility for the misuse of these tools and any unethical content created with them. This disclaimer is subject to the laws of the United States of America and the state of California. I reserve the right to remove or report any content that violates the terms of service, ethical standards, or applicable laws.<h2 id="heading-1489">License conditions</h2><a target="_blank" rel="ugc" href="https://civarchive.com/models/361666">Wagashi Style for Pony Diffusion v6 XL and Stable Diffusion 1.5 </a>© 2024 by <a target="_blank" rel="ugc" href="https://civarchive.com/user/pettankoutei">Pettan Koutei</a> is licensed under <a target="_blank" rel="ugc" href="http://creativecommons.org/licenses/by-nc-sa/4.0/?ref=chooser-v1">Attribution-NonCommercial-ShareAlike 4.0 International</a>

Wagashi Style

🔍 Version info:Initial release. Trained 100 images for 2800 steps.📋 Training Parameters:<ul><li>Base Model: NAI SD 1.5</li><li>Training Resolution: 576</li><li>Training Batch Size: 2</li><li>Type: LoCon</li><li>DIM / Alpha: 32 / 16</li><li>Conv Dim / Alpha: 24 / 6</li><li>Train On: Unet + TE</li><li>Optimizer: Prodigy</li><li>Scheduler: Cosine w/ restarts</li><li>Min SNR Gamma: 8</li><li>Bucket Resolution: 256 - 1024</li><li>Bucket Steps: 64</li><li>Noise Offset: Normal (0.035)</li><li>Image Count: 100</li><li>Epochs: 14</li><li>Repeats: 4</li></ul>

Files with SHA256: 798f61366e84e9d56ebdbb7925a7348ce0a3e98ffa11b008976e6c728a75b26c