<h2 id="style-lora-of-haoni-nihao-adx2kh97t">Style LoRA of <a target="_blank" rel="ugc" href="https://www.pixiv.net/en/users/1355685">HAONI / NIHAO</a></h2><ul><li>Like the style? Give the artist some support!</li></ul><h3 id="drop-a-and-follow-me-if-you-want-more!-8shjfdbpt">❤️ Drop a 👍 &amp; follow me if you want more!</h3><ul><li>I look forward to your posts, reviews and comments!</li></ul><h3 id="recommendations-0pwqd0jpi">🌟 Recommendations</h3><ul><li>To lessen the impact of Pony's inherent style<ul><li>Prompt: [score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up::0.2]</li><li>Negative: source_pony, source_cartoon, 3d, realistic, monochrome, censored</li></ul></li><li>I recommend using the model <a target="_blank" rel="ugc" href="https://civarchive.com/models/288584?modelVersionId=324692">AutismDPO </a>along with this <a target="_blank" rel="ugc" href="https://civarchive.com/models/152040?modelVersionId=360288">VAE</a>!</li></ul><h3 id="training-notes-zgirheeo6">📑 Training Notes</h3><ul><li>Do note that this LoRA is trained without score tags.</li><li>Please refer to the 'About this version' tab!</li><li>Model versions will be written in the following notation<ul><li>&lt;Config&gt;&lt;Config Version&gt;v&lt;Model Version&gt;</li><li>P2v1</li></ul></li></ul><h3 id="this-lora-can-only-work-on-pony-diffusion-v6-xl-or-models-derived-from-it.-05gpzc0ep">This LoRA can only work on <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749?modelVersionId=290640">Pony Diffusion V6 XL</a> or models derived from it.</h3>

HAONI / NIHAO Style (Pony XL)

📒 Notes<ul><li>Add 'sketch' to your prompt for a closer representation of the style!</li></ul>📈 Version Improvements<ul><li>Improved representation of style</li></ul>📑 Training Notes<ul><li>Base Model: <a target="_blank" rel="ugc" href="https://civitai.com/models/257749?modelVersionId=290640">Pony Diffusion V6 XL</a></li><li>Training Resolution: 1024</li><li>Training Batch Size: 4</li><li>Type: LoCon</li><li>DIM / Alpha: 16 / 8</li><li>Conv Dim / Alpha: 24 / 6</li><li>Train On: Unet</li><li>Optimizer: AdamW8bit</li><li>Scheduler: Rex</li><li>Min SNR Gamma: 5</li><li>Bucket Resolution: 512 - 2048</li><li>Bucket Steps: 64</li><li>Noise Offset: Normal (0.025)</li><li>Image Count: 242</li><li>Steps: 1300</li></ul>

Files with SHA256: 790a3c2cb0f842d43ba71f4e81a93af2e2f57d39bb7ba264aaa5081f5d20751a