I created this style using Hassaku as the base model, after that I only used the following prompt with the following models for testing:<ul><li>moonbean</li><li>Diamond</li><li>neverland</li><li>whimsical</li><li>hassaku</li><li>seasidemix</li></ul>Prompt for examples: AGGA_ST001, masterpiece, perfectly detailed, detailed face, detailed eyes, beautiful eyes, looking at viewer, 1girl, big breast, breast focuss<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4ae66a3d-a9a5-4393-81fc-51e60e6dd75d/width=525/4ae66a3d-a9a5-4393-81fc-51e60e6dd75d.jpeg" />As you can see I used a very simple prompt and it was enough to give me something decent n.n)/

Power_Mix

So... I was browsing youtube and found a video on how to merge loras sd1.5, although it was obviously very outdated, But Konya ss also allows merging loras sdxl so <a rel="ugc" href="https://civitai.com/articles/12860/sdxl-il-pdxl-or-test-lora-merging-in-google-colab">I updated it</a>... I'm learning how to use Google colab so I thought I'd upload my tests so if you like, you can try it out.

Files with SHA256: 92f510a90456f0307316db3efde3dbd478b36809c86847916f5246015e71d56e