this is a better version of my old virtual youtuber lora model train with more images of this is train with 1000 image of upper body and full body image of vtuber images of male and female vtuber. i recommend using it with <a target="_blank" rel="ugc" href="https://civarchive.com/models/55504/character-a-poses-vtuber-reference-pose-fullbody-halfbody">Character A-Poses / Vtuber reference pose (fullbody + halfbody) - v1.0 | Stable Diffusion LoRA | Civitai</a>i recommend using lora STRENGTH of 0.7 with better results only works for SD 1.5 anime type model only I'm still new to lora training

virtual youtuber

this version 2 is train on <a target="_blank" rel="ugc" href="https://civitai.com/models/85909/sudachi">Sudachi - v1.0 | Stable Diffusion Checkpoint | Civitai</a>

Files with SHA256: 74b7892765324c92fe6a2bd40a0ce92547fc3a8bd8a6a9df6494ceb877de3dab