Type | |
Stats | 401 260 |
Reviews | (62) |
Published | Apr 20, 2024 |
Base Model | |
Training | Steps: 3,682 Epochs: 14 |
Usage Tips | Clip Skip: 2 Strength: 1 |
Hash | AutoV2 64A58A0B59 |
该模型基于 Animagine XL V3.0 Base 底模训练
用法:
无需提示词触发。
推荐大模型:Animagine XL V3.1。
不使用 VAE。
采样器:Euler a,采样步数 28 步。
CFG Scale:6-7。
T2I 分辨率:不低于 768。
Lora 权重:1.0。
开启 ADetailer 可以更好地修复手和脸。
注意事项:
衣服在训练时作为训练内容,可能难以替换。
如果出现 NSFW 结果,请在正面提示词中添加 "jacket" 或更多文本/场景补充。
如果头发出现奇怪的颜色,请在负面提示词中添加 "multicolored hair"。
该 Lora 模型的训练数据集截取自番剧《GIRLS BAND CRY》第一集,采用标准优化器 AdamW 训练。欢迎使用其他模型(例如:animelllustDiffusion_v08)并分享你的成果。
This model is based on the Animagine XL V3.0 Base model.
Usage:
No prompt required.
Recommended base model: Animagine XL V3.1.
No VAE.
Sampler: Euler a, 28 sampling steps.
CFG Scale: 6-7.
T2I resolution: Not less than 768.
Lora weight: 1.0.
Enabling ADetailer can improve the generation of hands and faces.
Notes:
Clothes may be difficult to replace as they were part of the training data.
If NSFW results are generated, add "jacket" or more text/scene details to the positive prompt.
If hair has strange colors, add "multicolored hair" to the negative prompt.
This Lora model was trained on the first episode of the anime "GIRLS BAND CRY" using the standard optimizer AdamW. Feel free to use other models (e.g., animelllustDiffusion_v08) and share your results.