Trained on Animagine XL V3.1, 12 images.I'm bringing the SD LoRAs to the SDXL. I think you can notice she generates a bit different compared to the SDXL version, don't know why. I do know why, she's not trained on Hassaku, just saying she is going to generate a bit different. You can compare it with the old version, I'd say it's not a biggie.Another one of my failed projects. I'm happy I was able to bring her back, I was missing this type of secretary in my portfolio.I forgot if its her or Miyabi or both. Hands can generate poorly, but clothing can be a real pain. Let's just say the AI likes to generate her with cleavage or revealing clothing. Either try "()" to raise weight, or play around with wording if it bother you too much. Also, I spent like 30 tries to get her to pose with another character, but I don't know how to prompt it or what LoRAs to use, sad.I only used Animagine XV V3.1 to generate the images.I use Euler A. 34 steps. If you use PixAI or other currency sites, I recommend you go cheap first and then use more steps and stuff when you're sure you'll get the result you want.8CFGLoRAs I used in generation: <a target="_blank" rel="ugc" href="https://pixai.art/model/1725049259066326012">https://pixai.art/model/1725049259066326012</a> <a target="_blank" rel="ugc" href="https://pixai.art/model/1668692278399247652">https://pixai.art/model/1668692278399247652</a> <a target="_blank" rel="ugc" href="https://pixai.art/model/1826818504164548813">https://pixai.art/model/1826818504164548813</a>Thumbnail model: Animagine XV V3.1

Files with SHA256: 9c0b56184b42084cbb3aaf37316a59dde1a6f154e9343590177ee0f4b6356c3d