温馨警告
封面图为了骚气些是单独处理过的，只有头像框是用该模型生成，不用尝试该模型生成类似封面图，办不到。
一些症状说明
1. 目前1.0版本的模型出废图率还是比较高的，建议参考附带的例图写提示词。
2. 主要用于方形头像框生成，其他形状的未训练效果不理想。
3. 复制来的关键词出现“Y4PhF”时，需将其改为“头像框”（其他乱码显示的关键词都可删掉）
4. 该模型本身不输出明显的风格特征，有下列症状的都可尝试切换大模型或加入新LORA模型
整体风格不理想
某些局部图形或元素效果不理想，不还原，理解错误等
强烈建议搭配下方推荐的大模型使用
Description
FAQ
Comments (5)
This looks great!! 👏🤩
Are there english triggers as well for this LoRA? 🤔
(I'll likely give it a try later for fun anyway. Currently I'm busy. But thought it'd be worth asking for clarification for others, who might also be curious.)
“头像框”这是个伪触发词，可以不用输入；
主要看出图效果，如果AI输出的图在“头像框”的这个概念上表现较差时，可加上这个伪触发词，以达到强化“头像框”这一概念的目的。
@FangCM Okay! "Head-frame" might help, but this LoRA doesn't use any specific "trigger word." Just guiding the prompt into a format the Lora can utilize... 👍
伟大的！ 非常感谢~👌
-EDIT- Did some testing, it definitely wants that keyword. Plays quite nicely with RevAnimated, but strangely seems to struggle with just about any other checkpoint. Might be my settings though.
-EDIT2- Ah, looking over the samples provided again, I see now that there is quite a bit of guidance needed for other Checkpoints, but once applied tends to get more predictable results. Really appreciating these examples.
this looks inspired by the LPL art style , me like
We can make round frames ? For Discord profil
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.