Experimental Style Lora

这是一款实验性的画风LoCon。

This is an experimental style LoCon model.

使用方法 | Useage：

trigger word:

stylized

训练数据 | Train data：

由约100张AI生成的图片训练而成，包括了40+个人工选择的画面主题。

It was trained on around 100 AI-generated images, featuring over 40 manually selected visual themes.

生成配方：

step1:

Checkpoint: NoobAI-XL V-pred 1.0

CFG: 4.0 Sampler: Euler A

LoRa: Pony: People's Works + v7_NoobVv1.0 (strength:1.0) + STYLES | Illustrious/Noob RlI(Illust)0.1v (strength:0.6) + Cuteness Overload Style Illustrious XL v.1 (strength:0.4)

style prompt: masterpiece, best quality, very aesthetic, (realistic:0.9), (photorealistic:1), (flat color:0.5),

step2:

1.5 × upscale denoise: 0.2-0.3

LoRa: Pony: People's Works + v7_NoobVv1.0 (strength: 0.6) + 748cm | Style for Illustrious/NoobAI 0.75 v1.0 (strength: 0.6) + STYLES | Illustrious/Noob RlI(Illust)0.1v (strength:0.4)

style prompt: masterpiece, best quality, very aesthetic, (flat color:1),

这个操作由ComfyUI工作流完成。

These pictures are generated through a ComfyUI workflow.