[LoRA] Hason Game Icon Of Card

模型简介：
下个一个训练模型应该是宝箱，如果你有什么建议以及使用感受，可以给我留言,
希望 SXZ D.U.C.K. for Game Assets 的作者赶快更新，真的很好用！
▣ 建议使用权重为 1.3 左右
▣ 目前使用效果比较好的大模型有 Perfect World 完美世界 和 ReV Animated
▣ 建议生成了草图之后，使用图生图功能，会有较好的效果
▣ 可选择搭配lora SXZ D.U.C.K. for Game Assets 使用，权重建议0.3左右

The next training model should be a treasure chest. If you have any suggestions and usage experiences, please leave me a message,
I hope the author of SXZ D.U.C.K. for Game Assets can update it quickly, it's really useful!
▣ Suggest using a weight of around 1.3
▣ Currently, the most effective large models used include Perfect World 完美世界 and ReV Animated
▣ It is recommended to use the image generation function after generating a sketch, which will have better results
▣ You can choose to use it with Lora SXZ D.U.C.K. for Game Assets , with a recommended weight of around 0.3

~~~~~~~~~~~~~~~~~~~~~~~o(*￣︶￣*)o~~~~~~~~~~~~~~~~~~~~~~~

示例提示词和参数：
Prompt: 1girl,solo,closed mouth,night,starry sky,light particles,otonokizaka school uniform  ,pose, realistic,  medium breasts, beach,twilight,realistic, best quality,looking at viewer,

Negative prompt: (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), skin spots, acnes, skin blemishes, age spot, manboobs, backlight,anime, cartoon, drawing, illustration, boring,long neck, out of frame, extra fingers, mutated hands, monochrome, ((poorly drawn hands)), ((poorly drawn face)), (((mutation))), (((deformed))), ((ugly)), blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, glitchy, bokeh, (((long neck))), ((flat chested)), red eyes, extra heads, close-up, text ,watermark, logo,NG_DeepNegative_V1_75T, badhandv4

Steps: 20, CFG scale: 14, Sampler: DPM++ 2M Karras Test, Seed: 3997136979, ENSD: 31337, Size: 512x680, Model: majicmixRealistic_v2, Model hash: 5d22b29ecd, Hires steps: 14, Mimic scale: 10, Hires upscale: 1.4, AddNet Enabled: True, AddNet Model 1: HashimotoKanna(e57dbc81efd5), Hires upscaler: ESRGAN_4x, AddNet Module 1: LoRA, AddNet Weight A 1: 0.7, AddNet Weight B 1: 0.7, Denoising strength: 0.4, Threshold percentile: 100, Dynamic thresholding enabled: True, Clip skip: 2