
Trained on over 100 screenshots and promo art drawings from the original series. There's already several Loras based on the more detailed reboot design, but I think there's also a charm to the more streamlined cartoony angular characters of the first show as well.
As Anna switches up her scarf location from her head to her neck midway through, you'll want to specify whether it's on top of her head or around her neck if you want to include it. Some models are a bit finicky with her eye color and default to red or blue, but I find emphasizing commands like dark/grey/grey-black eyes helps.
I've already cooked a Flux equivalent on this same dataset, but need to both get a better hang of the right commands for it and generate more good sample pics before publishing.
描述:
训练词语: anna2k1,young girl,sleeveless black dress,red scarf,necklace,sandals,dark eyes,blonde-brown hair,slender
名称: 917945_training_data.zip
大小 (KB): 22978
类型: Training Data
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: anna2k1_pony.safetensors
大小 (KB): 223110
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success