I continue to try to train LoRA locally. This LoRA generates images of girls in a Megumin costume. Sometimes it gives out poorly drawn hands, sometimes pixelated faces. I managed to select prompts for high-quality generation, you can see the examples. I achieved the best results with <a target="_blank" rel="ugc" href="https://civarchive.com/models/101055?modelVersionId=126601">Base SDXL</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/192854?modelVersionId=1813193">Ratatoskr</a> models, but <a target="_blank" rel="ugc" href="https://civarchive.com/models/139562?modelVersionId=789646">RealVis</a> sometimes not bad too. I used str 1, cfg 6.0, dpmpp_2m_sde, karras

Files with SHA256: 2c10f8cc45c650b220b004a9a11fda07167ea37292fe0a0cdc054e182e6b9be7