Makes heavily stylized images similar in style to the artist <a target="_blank" rel="ugc" href="https://gelbooru.com/index.php?page=post&amp;s=list&amp;tags=channel_%28caststation%29">Caststation</a>.How to use This LoRA was trained entirely on screencaps from the viral anime music video as a test on making general style models from low variation datasets. It came out surprisingly versatile and works great to make outputs with thick linework in limited palettes. Recommended LoRA strength is between 0.7 yo 1.0. At higher levels you may find problems prompting school uniforms in different designs from the one in the video. I also recommend using a colored or "simple background" for best results. For those looking to prompt smuggies the tags "smug", "open mouth", "half-closed_eyes" work well together.MISC Trained on Pony XL with a dataset of 86 images. All images are raw txt2img upscaled, no inpainting or helper embeddings used for quality transparence, e.g. - things should look even better if you do. The training was done on a small subset of images, as such the output of this LoRA is not an accurate representation neither in quality or style of the source material. Go support the <a target="_blank" rel="ugc" href="https://x.com/x_cast_x/">artist</a><a target="_blank" rel="ugc" href="https://skeb.jp/@yocto_mako"> </a>if you want the good stuff.

Files with SHA256: 1b8b1b9d6ce7810c6c25703769c79221d501fb38684ce014959d0358f3a93cb2