<h1 id="for-sdxl-version-m60975pni">FOR SDXL VERSION</h1>Trained on the <a target="_blank" rel="ugc" href="https://civarchive.com/models/277058?modelVersionId=1346244">epiCRealism XL</a> model and should therefore be used with this model for perfect use.I'm looking for photographic rendering only, I'm not interested in 3D or 2D rendering. So my Lora will probably never be trained on base model versions.<ul><li>I highly recommend using After Detailer.</li><li>Trigger token is ohwxtd</li><li>Can be used at an intensity of 1.0 but I always prefer to stay at 0.9</li></ul>Info about train settings :<ul><li>LoRA trained with a dataset of 107 images / 10 repeats / 10 epoch with batch size at 1.</li><li>Mixed and save precision = bf16</li><li>Optimizer = AdamW</li><li>Resolution = 1024x1024px</li></ul><h1 id="for-sd-1.5-version-94gmsljz0">FOR SD 1.5 VERSION</h1>Trained on the <a target="_blank" rel="ugc" href="https://civarchive.com/models/132632?modelVersionId=201259">epiCPhotoGasm-Z</a> model and should therefore be used with this model for perfect use.But work also very well with <a target="_blank" rel="ugc" href="https://civarchive.com/models/97744/realcartoon-realistic">RealCartoon-Realistic</a>I'm looking for photographic rendering only, I'm not interested in 3D or 2D rendering. So my Lora will probably never be trained on base model versions.<ul><li>I highly recommend using After Detailer.</li><li>Trigger token is ohwxtd</li><li>Can be used at an intensity of 1.0 but I always prefer to stay at 0.9</li></ul>Info about train settings :<ul><li>LoRA trained with a dataset of 65 images / 8 repeats / 13 epoch with batch size at 4.</li><li>Mixed and save precision = bf16</li><li>Optimizer = AdamW </li><li>Resolution = 512x512px</li></ul>

Thaïs d'Escufon : Realistic version.

Improvements to v1: <ul><li>More steps (2500)</li><li>More pics in dataset (96)</li></ul>

Files with SHA256: e1cc7e51bf513e62cc6ecbf4a5842e02057a19d7ca9c87da8aec2ba0cd4ee568