WooMix (Illustrious)

详情

下载文件

关于此版本

1. hassakuXLIllustrious_v13StyleB

2. hassakuXLIllustrious_v32

3. prefectIllustriousXL_40

4. oneObsession_v18

I then performed a series of merges:

I first blended models *1 and 2** together at a 65/35 split.

Then I blended models *3 and 4** together at a 60/40 split.

Finally, I merged the results of those two previous blends together at a *50/50 ratio**.

After getting that base blend, I layered on two specific LoRA files, both at a very low influence setting of 5% strength:

1. The LoRA named USNR_STYLE_ILL_V1_lo...

2. The LoRA named 748cm_illu.safetens...

I blended the outputs of applying these two LoRAs together one last time at a 50/50 ratio.

For the very last piece, I used the VAE named purecolorVAE_v10.safetensors to finalize the colors.

F seems to have less black outlines in testing

模型描述

我认为使用 CFG 5 可以达到完美的饱和度和锐化效果。

尝试使用以下负面标签：“worst quality, low quality, lowres, blurry, pixelated, jpeg artifacts, oversaturated, watermark, signature, unfinished, sketch”，有时减少负面标签反而能获得更好的图像，甚至可以使用：“worst quality, low quality, lowres, blurry, pixelated, jpeg artifacts, mosaic blur, watermark, signature, unfinished, sketch, bad hands, distorted eyes, bad fingers”。

现在尝试使用：“worst quality, low quality, lowres, jpeg artifacts, cutoff, grainy, distorted, watermark, error, unfinished, sketch, draft, deformed, poorly drawn face, poorly drawn eyes, poorly drawn hands, fused fingers, bad hands, deformed hands, bad fingers, distorted eyes, censored, mosaic censoring, futanari, monochrome”，如果你不喜欢，可以删除 futanari。

或者甚至只用：“worst quality, low quality, lowres, bad”，更好的是：“worst quality, error”，因为过多的负面标签可能会让 AI 混乱。

此模型生成的图像

排序

未找到图像。

模型类型	检查点
基础模型	Illustrious
发布时间	10/26/2025