MikuIllustrious1 / README.md
M5Dev's picture
Create README.md
cf6593e verified
---
language: en
license: creativeml-openrail-m
base_model: OnomaAIResearch/Illustrious-XL-v2.0
tags:
- stable-diffusion
- text-to-image
- lora
- anime
- hatsune-miku
- vocaloid
- illustrious-xl
inference: true
pipeline_tag: image-to-video
---
# MikuIllustrious1 - Hatsune Miku LoRA for Illustrious-XL v2.0
**The ultimate Hatsune Miku LoRA** trained on **Illustrious-XL v2.0** using high-quality dataset. Perfect for generating beautiful, detailed, and seductive Miku in various scenes (onsen, bedroom, stage, etc.).
### Trigger Word
Use **`MikuIllustrious1`** at the beginning of your prompt to activate the LoRA.
### Recommended Settings
- **Base Model**: Illustrious-XL v2.0 (or any compatible v2.0 variant)
- **LoRA Weight**: 0.85 ~ 1.0 (0.9 is usually the sweet spot)
- **Steps**: 28-40
- **CFG Scale**: 6-7
- **Sampler**: Euler a / DPM++ 2M Karras
- **Resolution**: 1024x1536, 1216x832, 1024x1024 (portrait works best)
- **Clip Skip**: 2 (if using ComfyUI: CLIPSetLastLayer -2)
### Notes
- Best results when using **Illustrious-XL v2.0** base model.
- The LoRA was trained with a strong focus on wet skin, dynamic lighting, hair details, and seductive/cute expressions.
- Works excellently in ComfyUI, Forge, and PixAI.
- Supports DiT.2 architecture.
### License
This LoRA follows the **CreativeML OpenRAIL-M** license.
You are free to use it for both personal and commercial purposes, but please credit the author if possible.
**Made with love by M5 Dev** 💚
If you like this LoRA, feel free to leave a like or comment! More versions coming soon 🔥