Hello, I'm here to share my Vtuber AI project! Basically, this is a project I started at the end of 2025 when I was learning C++ and decided to create my own AI using this programming language. Since then, I've been working hard on it to start doing live streams. The main idea was to create a mascot that would attract people to a forum I'm still going to create, and thus be a good way to promote it. Currently, the project runs basically 90% locally, even though I have a very modest PC. I don't want to go into too much detail so that the post doesn't get moved to /t/, but I recently completed the first version of this project and I would like to ask for your opinion. The only non-local part that runs is its connection to Twitch, where it reads Twitch messages and responds to those who type "himiko". Regarding its visual model, I have a 2D model, a VRM model, and a model that I tried to animate in Inochi Session, but I didn't want its camera movement to come from mine. Therefore, I only used her 2D model that appears when she speaks. If anyone knows how to work with motion animation for vtubing, I still have her Krita project file! If you're interested in this project, I can post a detailed explanation of how it works on /t/, since I created a mini emulator of emotions and emotional states for her. Her introductory video is this one here. https://www.youtube.com/watch?v=7irb4qkKJJc