动漫风景/背景/插图Pony版

使用介绍

这个实验性 LoRA 旨在通过 Pony Diffusion 检查点生成详细的动漫风格背景。

可用于生成背景艺术，并在提示角色主题时减少简单或白色背景的出现。

（simple_background、white_background在booru角色训练数据中很常见）

v2 中的新功能：

该版本在AutismMix SDXL上进行训练

除了常规的 Pony分数修饰符之外，美学质量修饰符标记现已应用于训练数据，包括在提示末尾：

masterpiece, best quality, very aesthetic, absurdres

*这是正在进行的实验的一部分，旨在将 Pony 扩散微调到质量标签，使其能够与其他 SDXL 模型顺利合并。

提示结构

正面提示（注意提示末尾的质量修饰符）：

score_9, score_8_up, score_7_up, score_6_up, source anime, scenery,
{{natural language description}}
{{tags}}
masterpiece, best quality, very aesthetic, absurdres

负面提示：

score_4, score_5, score_6, monochrome, censor, 3d, render, CGI, text, watermark, cartoon, sketch, painting, impasto,

触发词：	SCENERY MASTERPIECE BEST QUALITY VERY AESTHETIC
使用提示：	CLIP SKIP: 2 STRENGTH: 0.8

封面提示

score_9, score_8_up, score_7_up, score_6_up, source anime,
a girl standing on a train track, tree, nature, forest, bush, grass, 1girl,
masterpiece, best quality, very aesthetic, absurdres,
<lora:detailed_backgrounds_v2:1>
Negative prompt: score_4, score_5, score_6, monochrome, censor, 3d, realistic, render, CGI, text, watermark, cartoon, sketch, painting, impasto,
Steps: 30, Size: 768x1344, Seed: 319733556, Model: autismmixSDXL_autismmixPony, Version: f0.0.17v1.8.0rc-latest-276-g29be1da7, Sampler: Euler a, CFG scale: 7, Clip skip: 2, Model hash: 821aa5537f, Hires upscale: 1.4, Hires upscaler: Latent, ADetailer model: person_yolov8n-seg.pt, ADetailer prompt: [object Object], ADetailer version: 24.4.1, Denoising strength: 0.6, ADetailer mask blur: 4, ADetailer model 2nd: face_yolov8n.pt, ADetailer confidence: 0.3, ADetailer prompt 2nd: [object Object], ADetailer dilate erode: 4, ADetailer mask blur 2nd: 4, ADetailer confidence 2nd: 0.3, ADetailer inpaint padding: 32, ADetailer dilate erode 2nd: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding 2nd: 32, ADetailer denoising strength 2nd: 0.4, ADetailer inpaint only masked 2nd: True

关联模型