<h1 id="obligatory-disclaimer:-this-lora-is-intended-to-be-used-alongside-style-loras-andor-artist-style-tags-to-bring-out-its-full-potential.-all-example-images-use-wai-anima-and-the-&quot;haku89&quot;-artist-style-tag.">Obligatory Disclaimer: This LoRA is intended to be used alongside style LoRAs and/or artist style tags to bring out its full potential. All example images use WAI-Anima and the "haku89" artist style tag.</h1><h2 id="if-you-don't-see-all-4-example-images-try-changing-browsing-level-to-&quot;r&quot;-or-lower.">If you don't see all 4 example images, try changing Browsing Level to "R" or lower.</h2><h3 id="comments-and-gallery-posts-are-encouraged-and-appreciated-regardless-of-their-content!-(within-tos-of-course.)">Comments and gallery posts are encouraged and appreciated regardless of their content! (Within ToS, of course.)</h3>Start with a model strength of 1.0. If you find the model to be too stiff, try lowering it to 0.80.A character LoRA for the indie TikTok VTuber Nabins. Trained on a dataset of 40 images that consists largely of cut up screenshots, images generated by ChatGPT, and a few fan arts.I've changed my LoRA training settings once more in hopes of better flexibility. I've swapped back to Anima Base since I think the generations look marginally better and background generation appears to be significantly nicer. I've kept the halved step count Prodigy optimizer though. That being said, it's still not perfect and the are definitely still issues. Most notably, the model is very hesitant to generate her jacket completely unzipped at a model strength of 1. Lowering the strength to something like 0.80 appears to generate a completely unzipped jacket more often. Proper weighting and more detailed natural language prompted can also very likely fix this. There's also some melting with the very tips of the hair but this is very minor and not really visibly unless zoomed in. The below and attached triggers are general suggestions. You can almost certainly get better and more consistent results by more specific prompting. I'm just lazy. Play around with it to see what works for you. Remember, Anima loves natural language so converting the listed triggers into natural language can result in more consistent generation.Detailers are highly recommended for the face and eyes.Manual inpainting or image refinement through ChatGPT are recommended if you're looking to generate small details.Probably won't be updated.All example images were generated without a refiner step. You can probably make your generations look much better with improved workflows.Triggers:Nabins, short hair, two-side up, pink hair, pink eyes, large breasts, cat ears,white jacket with dark pink sleeves, fluffy white slippers,

Files with SHA256: 733fd05cecd973d22739889f6fdf1c7828088089b47b2d95c27b2042325c77bf