<h2 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.</h2><h3 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.</h3>A character LoRA for the indie VTuber SoraVA (also know as Sora Reacts, Sora Anime, and Sora ASMR) who uses the premade Cuckoo ASMR model. Trained on a dataset of 40 images that consists largely of cut up screenshots, images generated by ChatGPT, and a few fanarts.I'm still very new to Anima LoRA training so do forgive any issues or weirdness. As such, this LoRA does have its quirks. The main issue is that it has a tendency to generate her in a kneeling position due to the Cuckoo ASMR model completely lacking a standing pose. This was such an issue that I had to retrain using a few synthetic images that depict her standing. It does still sometimes default to kneeling but this can now be much more easily fixed with weighting or descriptive natural language prompting.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:SoraVA, long hair, heart-shaped ahoge, purple eyes, large breasts, single wing on the side of head, white waist wings, frilly white collar, white drop-like necklace, see-through white frilly and lacy dress, short frilly skirt, detached white frilly sleeves, white arm wrapping, white thighhighs, no shoes,

Files with SHA256: 0d0e59e2878a804e39f61f49034c87be893926fda1ce36a496bd9b5d6918ff17