Dive into the enchanting world of Ixy's distinctive art style with the Ixy Style LoRA!This LoRA meticulously captures the signature aesthetic of the renowned Japanese artist Ixy (いくしー). Known for his exceptionally cute and charming anime girl illustrations, Ixy's style is characterized by clean, crisp line art, vibrant yet often flat coloring, and an undeniable sense of expressiveness. This model excels at reproducing these elements, bringing a delightful blend of simplicity and detailed character focus to your generations.Whether you're looking to recreate the adorable essence of Ixy's original works or apply his signature touch to your favorite characters, this LoRA is your gateway to stunning, high-quality illustrations.<hr /><h3 id="key-features:">Key Features:</h3><ul><li>Characteristic Kawaii: Generates characters with Ixy's quintessential 'kawaii' appeal, featuring large expressive eyes, adorable expressions, and a generally youthful, charming look.</li><li>Clean &amp; Crisp Aesthetics: Emphasizes clear line work and a generally flat, appealing color palette, reminiscent of professional digital illustrations.</li><li>Focus on Portraiture: Optimized for upper-body and character portrait compositions, making it perfect for profiles and expressive character shots.</li><li>Signature Backgrounds: Often produces images with simple or clean white backgrounds, allowing the character to truly stand out, faithful to Ixy's common style.</li><li>Versatile Character Application: While trained on a diverse set of Ixy's beloved illustrations (including various characters from Azur Lane and other well-known anime series), this LoRA can be effectively applied to new characters, allowing you to reimagine your favorite figures in Ixy's unique style.</li></ul><hr /><h3 id="recommended-usage:">Recommended Usage:</h3><ul><li>Trigger Word: Always use ixy in your positive prompts.</li><li>LoRA Strength: A LoRA strength of 0.7 to 1.0 is generally recommended for optimal results, though feel free to experiment based on your desired level of style adherence.</li><li>Base Model: This LoRA was specifically trained on Illustrious v0.1 and performs best with it, or other anime-focused models compatible with Illustrious v0.1. (It supports danbooru tags as well!)</li><li>Positive Prompts:<ul><li>Incorporate descriptive tags that reflect Ixy's style: <code>cute, anime style, detailed hair, detailed eyes, upper body, white background, simple background, digital art, character portrait</code>.</li><li>For specific looks, add common danbooru tags like <code>pink hair, blue eyes, blush, crying, jacket, off-shoulder shirt, cat ears, military uniform</code>, etc.</li><li>Stylize any character in danbooru u want, using the danbooru tag like <code>oyama mahiro</code>.</li></ul></li><li>Negative Prompts: Use your preferred standard negative prompts to combat common issues like bad anatomy, blurry images, distorted features, etc.</li><li>Sampling Method: DPM++ 2M Karras, DPM++ SDE Karras, Eular A</li><li>Resolution: Start with 832x1216 for best fidelity, and then upscale if needed.</li></ul>

ixy（いくしー）

This is an incremental update focusing on refining the ixy style reproduction.Key Updates in this Version:<ul><li>Enhanced Style Stability: Through minor adjustments to the learning rate and an decreased training steps, the model exhibits more consistent and stable adherence to the core ixy art style.</li><li>Dataset Refinement: Suboptimal images were removed from the training dataset, contributing to cleaner outputs.</li></ul>Known Limitations:While overall style consistency is improved, we've observed occasional instances of hand deformities or slight color bleed/overflow in some generations. We are actively working on addressing these minor artifacts.Looking Ahead:Future updates will focus on further refining dataset tags to mitigate the identified issues and enhance overall quality. Your feedback and generated images are invaluable for our continuous improvement!

ixy-illustrious-v-01-3-000006.safetensors

v1.1(IllustriousXL-v0.1)

v1.0(IllustriousXL-v0.1)

### 当前模型的贡献者未提供更加详细的模型介绍。模型文件和权重，可浏览“模型文件”页面获取。
#### 您可以通过如下git clone命令，或者ModelScope SDK来下载模型

SDK下载
```bash
#安装ModelScope
pip install modelscope
```
```python
#SDK模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('ModelE/ixy-illustrious-v-01-3')
```
Git下载
```
#Git模型下载
git clone https://www.modelscope.cn/ModelE/ixy-illustrious-v-01-3.git
```

如果您是本模型的贡献者，我们邀请您根据<a href="https://modelscope.cn/docs/ModelScope%E6%A8%A1%E5%9E%8B%E6%8E%A5%E5%85%A5%E6%B5%81%E7%A8%8B%E6%A6%82%E8%A7%88" style="color: lightgrey; text-decoration: underline;">模型贡献文档</a>，及时完善模型卡片内容。

Files with SHA256: 4849704f8674c58c1f24db51b02973f24ef1272151ce48ea0e65f38368c1a650