lora说明<ul><li>每一张画面都会被各种形状的框架约束着。 | Each picture is constrained by frames of various shapes.</li><li>远景、中景、近景都有不同的层次。 | There are different levels of far, middle, and near scenes.</li><li>颜色丰富，但是更倾向节日色调。 | The colors are rich, but they tend to be festive colors.</li><li>男孩、女孩、动物、植物、建筑都可以很好的支持。 | Boys, girls, animals, plants, and buildings can all be well supported.</li></ul>使用说明<ul><li>大模型推荐选择万向熔炉等没有经过过多融合的模型。 | Large models are recommended to choose universal melting pot and other models that have not undergone too much fusion.</li><li>触发词是zhipianren。 | The trigger word is zhipianren.</li><li>初次使用可以尝试输入 中国建筑、中国服饰、云，太阳、月亮，等提示词可以更快的得到和示例图相似的画面。 | For the first time, you can try to input prompts such as Chinese architecture, Chinese clothing, clouds, sun, moon, etc., to quickly get a picture similar to the example picture.</li><li>可以配合ControlNet来生成文字或者控制画面构图。 | It can be used with ControlNet to generate text or control the composition of the picture.</li><li>权重推荐使用1-0.8。 | The weight is recommended to be used between 1 and 0.8.</li></ul>

Files with SHA256: 8f3bd8adb6c94eb617344aa27c8143f8d97926cba0f76e9819fc4a06d9741d41