<p><a target="_blank" rel="ugc" href="https://danbooru.donmai.us/artists/623527">Artist page</a></p><div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/UKcJqQqiXq0" start="0"></iframe></div><p></p>

@kyuhirano 平野九 / 無職転生3 ED 「祈り、終われば」 Style

<p>Initial.<br /><br />The LoRA was trained on 96 frames extracted from the ending promotional video (ED PV).</p><p>Captions were generated using <code>SmilingWolf/wd-eva02-large-tagger-v3</code> with a threshold of <code>0.35</code>. Incorrect tags such as <code>colored_skin</code> were removed, and character names, camera angles, and other essential details were minimally supplemented.</p><p>No natural-language captioning was performed. However, natural-language descriptions in prompts may still evoke compositions from the training images.<br /></p><p>LoKr, dim32/alpha32, CAME, UNet LR 3e-05 cosine 1,000steps, 96 images (512² res).</p>

Files with SHA256: 6bf279f649fbcf7f4a08f758f8228ab5a1dcb7713212008116512f0ad6c9525a