Kimiko is my first fully fictional character created with AI. I made this LoRA to explore a highly versatile character that can fit into many different concepts, starting with a first training on Z-Image Turbo.There are still a few limitations for now: her hairstyle is hard to change, her hair often looks wet, and her outfit tends to default to suspenders, mostly because the dataset is still limited. On the positive side, I’m happy with her expressiveness, the realistic textures, and her overall cute look.If I release this model, it’s with the hope that your creations will help enrich the next dataset and shape an improved version of this LoRA. You’re free to use her creatively, but please avoid using her face for commercial purposes.My plan is to keep improving her for Z-Image Turbo first, then expand her to other models such as Flux2Klein or Chroma as photorealism continues to evolve. I hope you enjoy Kimiko, and if this project works well, I’ll probably create more characters in the future. If you like this LoRA, don"t hesitate to:<ul><li><a target="_blank" rel="ugc" href="https://emojipedia.org/thumbs-up">👍 l</a>iking this model and <a target="_blank" rel="ugc" href="https://civarchive.com/user/LOREVER">follow me</a></li><li>publishing your photos directly in the model's gallery.</li><li><a target="_blank" rel="ugc" href="https://ko-fi.com/lorever">support me</a> by giving some <a target="_blank" rel="ugc" href="https://civarchive.com/images/46221635">BU</a><a target="_blank" rel="ugc" href="https://civarchive.com/images/26283008">ZZ</a> (to train next models)</li><li><a target="_blank" rel="ugc" href="https://tensor.art/u/823163966116296190/models">visit my Tensor account</a> to try most of my models.</li></ul><h3 id="thank-you-to-everyone-who-supports-me-!any-resemblance-to-real-persons-living-or-dead-is-purely-coincidental.">Thank you to everyone who supports me ! Any resemblance to real persons, living or dead, is purely coincidental.</h3>

Kimiko - Fictional Character

Works well at 0.9 - 1.0 Known issues: Hair (style and wet effect), clothes (thin straps), Lack of context (default grey background), gaze (often to the left). I'll improve that in V2

Files with SHA256: 9268a6f44dcc09ee047da279a3cb33fb0e42003428fb05690059a4d19227b8b5