This LoRA was trained on both the Arcana version and the base Queen of Pain model. If you want to generate only the base version of the character, add "Queen_Arcana" to the negative prompt.

The dataset also contains some futanari versions of the character. If you want to avoid generating those features, try adding "futanari" and "penis" to the negative prompt.

For some reason, the anima-based model wouldn't let me use as many training steps as I wanted, so the final result ended up a bit strange. It's still usable, but it's not quite where I'd like it to be in terms of quality and consistency. I'll probably revisit it in the future and retrain it with better settings if I find a way around the limitation

Please leave feedback and share your creations. Seeing what people generate is not only enjoyable, but also extremely helpful for improving the model. Your images, prompts, and feedback help identify the model's strengths and weaknesses, making future updates more effective.

If you encounter issues, have suggestions, or would like to see specific improvements, features, outfits, styles, or additional versions of the character, let me know. Future updates to both the LoRA and its description may be influenced by your feedback and the work you share.