This is a lora trained on the images of a pixiv user named matcha_
I have a checkpoint version of this which is pretty bad so this version is the better version and upgrade and has the benefits of the LORA
matchaStyleLora_v1.safetensors
抹茶matchaStyleLora_v1.safetensors
MatchaStyleV1.safetensors
Same model published on other platforms. May have additional downloads or version variants.