Model Introduction: This is a "light and shadow type Lora" that can give your image a light and shadow effect of light mint green. It comes with a depth of field, background blurring, and lower saturation in the generated image, giving it a fresh, warm, and hazy beauty. Inspired by "Dream of Rouge Lora"
Big Model: MajicMIX Realistic Maiju Realist_v7-fp16.safetensors
Recommended sampling steps: 20-30, size 512 * 768
Trigger word: chunguang, glossy, green theme, (Trigger words can also be deleted as appropriate, and adding them all can have a more prominent effect. In most cases, not using trigger words can also have a very good effect.)
Due to the large number of portrait characters in the training set, not specifying the direction of the light source when using a real person model to create images will result in inherent light emission, which can also affect the precision of the face and hands in the base model. For example, when drawing panoramic portraits, if hand movements are not specified, there is a chance of producing camera decorations. If clothing is not specified and green themes and bikinis appear, the weight can be appropriately reduced or layered for use.
Model recommendation weight is 0.5-1.0. If you like it, you can try it more according to different large models
Suggest using it to describe beautiful scenes or outdoor scenery (skirts, grass, flowers, forests, ponds, lakes). And this Lora is suitable for various realistic, 2D, and 2.5D large models, often paired with some obscure models to have surprising effects!
The author said: Welcome to return to the picture for mutual assistance and encouragement. If you have any suggestions, please leave a message
If you have any model customization or other cooperation or consulting needs, please contact VX: 1weiya-12138. Please briefly indicate your purpose. Time is precious, please do not chat. Thank you for your understanding!
Basic Algorithm 1.5
Scope of Creative License
Can generate images online
Downloadable
Can be fused
The map can be shared and released on the Internet platform
Business License Scope
To sell generated images or use them for commercial purposes, one needs to consult oneself for authorization
Non resale model or sale fusion model
*The permission scope is set by the creator themselves, and users are required to use it in accordance with regulations
Enterprises using models for commercial purposes can contact the platform for consultation on model commercial authorization
——————————————————————————————————————————
模型简介：这是一款【光影类lora】，可以让你的画面有淡淡薄荷绿的光影效果，自带景深，背景虚化，生成图会有较低的饱和度，有种清新温暖朦胧的美感，灵感来源：胭脂梦Lora
大模型：majicMIX realistic 麦橘写实_v7-fp16.safetensors
采样步数推荐20-30 尺寸512*768
触发词：chunguang,glossy, green theme, （触发词也可以酌情删减，全加上效果更突出，大部分情况下不用触发词也有很不错的效果）
由于训练集里有大量人像角色，在使用真人模型出图时不指定光源方向会自带发丝光，也会对底模的人脸以及手部精细度产生影响，如画全景人像如不规定手部动作有几率产出相机装饰，如不规定服装会出现绿色主题和比基尼，可以适当降低权重或者分层使用,
模型推荐权重0.5-1.0，喜欢的伙伴可以根据不同大模型多试试
建议搭配形容比较唯美的场景或者户外风景使用（裙子、草地、花、森林，池塘,湖水）,并且此Lora适应各种写实、2D、2.5D大模型，往往配合一些冷门模型会有惊喜的效果！
作者说： 欢迎返图 互助共勉 如有建议 欢迎留言
如您有模型定制或其他合作、咨询需求，请联系VX: 1weiya-12138 请简单注明来意，时间宝贵不闲聊哈，谢谢理解！
基础算法 1.5
创作许可范围
可在线生图
可下载
可进行融合
生图可以用于分享，互联网平台发布
商业许可范围
出售生成图片或用于商业目的需咨询本人获取授权
不可转售模型或出售融合模型
*许可范围为创作者本人设置，使用者需按规范使用
企业将模型用于商业用途，可联系平台咨询模型商业授权
描述:
训练词语: chunguang,glossy, green theme,
名称: 春光-0.3.safetensors
大小 (KB): 36978
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
