Bringing a character to your side from nothing more than a single illustration—that's the vision we set out to build. She blinks, turns her head, shifts expressions, and reacts to your touch. Nothing in this video is hand-keyed; every single movement is running live in our engine. Meet Iroha Wakana, the face of our project, whose signature line is "…Not this one." No rigging skills required, and no slicing your art into layers by hand. Just drop in one illustration and watch your character come alive. This is our first look—upper body only for now—with plenty of details still being refined. We aren't here to claim we've dethroned Live2D. We still have a ways to go to reach that bar, but today marks a massive leap forward. Deepest thanks to the two open-source projects that laid the foundation for Iroha: • See-through by shitagaki-lab (ljsabc): splits an illustration into layers github.com/shitagaki-lab/see… • Anime2.5DRig by 8co28: brings layered art to life in the browser github.com/852wa/Anime2.5DRi… We've built extensively on top of both to push this even further. Her voice is synthesized via GPT-SoVITS, using reference audio from Amitaro's Voice Material Studio (amitaro.net). Launching soon. If you have any questions, leave them below—we'll reply to as many as we can. #Live2D #anime #animeart #OC #originalcharacter #illustration #VTuber #animation #animegirl #indiedev