---
title: Anycoder 5f028fd4
emoji: 🏢
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 6.6.0
app_file: app.py
pinned: false
tags:
- anycoder
---
=== README.md ===
```markdown
---
title: 🎨 Magic Wan Image V2 - Text to Image
emoji: 🎨
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 6.0.2
app_file: app.py
pinned: false
license: apache-2.0
tags:
  - text-to-image
  - diffusers
  - wan
  - lora
  - image-generation
  - ai-art
  - gradio
---

# 🎨 Magic Wan Image V2 - Text to Image Generator

Generate stunning images from text prompts with **Real Wan LoRA Support**! This Space uses the [Magic-Wan-Image-V2](https://huggingface.co/wikeeyang/Magic-Wan-Image-V2) model with support for various Wan-compatible LoRA adapters.

<div align="center">
  <a href="https://huggingface.co/spaces/akhaliq/anycoder">
    <img src="https://img.shields.io/badge/Built%20with-anycoder-FF9D00?style=for-the-badge" alt="Built with anycoder">
  </a>
</div>

## ✨ Features

- **🎭 Multiple LoRA Styles**: Choose from 15+ Wan-compatible LoRA adapters including:
  - Anime/Manga Style
  - Realistic Vision
  - Oil Painting
  - Watercolor
  - Pixel Art
  - 3D Render
  - Sketch
  - Cyberpunk
  - Fantasy
  - Studio Ghibli
  - Vintage Photo
  - Pop Art
  - Minimalist
  - Horror
  - Concept Art

- **⚙️ Full Control**: Adjust all generation parameters:
  - Width & Height (256-1024px)
  - Inference Steps (10-100)
  - Guidance Scale (1.0-20.0)
  - Seed (with randomization option)
  - LoRA Scale (0.0-2.0)

- **📚 Example Prompts**: Click pre-made examples to quickly test different styles

- **🚀 Optimized Performance**: FP16 inference with CUDA acceleration when available

## 🚀 How to Use

1. **Enter your prompt**: Describe the image you want to generate in the Prompt box
2. **Optional negative prompt**: Specify what to avoid in the generated image
3. **Select a LoRA style**: Choose from the dropdown menu (or keep "None" for base model)
4. **Adjust settings**: Modify LoRA scale, dimensions, steps, and guidance as needed
5. **Click Generate**: Wait for your image to be generated!

### 💡 Tips for Better Results

| Parameter | Recommendation |
|-----------|----------------|
| **Prompt** | Be descriptive and specific. Include style, lighting, composition details |
| **LoRA Scale** | Start at 1.0, increase for stronger style effect (max 2.0) |
| **Steps** | 30 steps is a good balance. More steps = better quality but slower |
| **Guidance** | 7.5 is default. Higher = more prompt adherence, lower = more creative |
| **Seed** | Use same seed for reproducible results, or randomize for variety |

### 📝 Example Prompts

```
A cyberpunk city at night, neon lights, rain, highly detailed, 8k
Portrait of a fantasy elf warrior, intricate armor, forest background
Magical library with floating books, mystical atmosphere, warm lighting
Steampunk airship flying through clouds, detailed mechanical parts
```

## 🏗️ Technical Details

### Model Information

- **Base Model**: [Magic-Wan-Image-V2](https://huggingface.co/wikeeyang/Magic-Wan-Image-V2)
- **Framework**: Diffusers + PyTorch
- **Precision**: FP16 (for faster inference)
- **LoRA Support**: Full Wan-compatible LoRA adapter support

### LoRA Adapters

All LoRA adapters are automatically downloaded from Hugging Face Hub on first use and cached locally. The following repositories are used:

| LoRA Name | Repository |
|-----------|------------|
| Anime Style | AlienKevin/Wan2.1-LoRA-Anime |
| Realistic Vision | prithivMLmods/Wan-Realistic-LoRA |
| Oil Painting | strangerzonehf/Wan2.1-LoRA-Oil-Painting |
| Watercolor | strangerzonehf/Wan2.1-LoRA-Watercolor |
| Pixel Art | strangerzonehf/Wan2.1-LoRA-Pixel-Art |
| 3D Render | prithivMLmods/Wan-3D-Render-LoRA |
| Sketch | strangerzonehf/Wan2.1-LoRA-Sketch |
| Cyberpunk | prithivMLmods/Wan-Cyberpunk-LoRA |
| Fantasy | prithivMLmods/Wan-Fantasy-LoRA |
| Studio Ghibli | AlienKevin/Wan2.1-LoRA-Ghibli |
| Vintage Photo | strangerzonehf/Wan2.1-LoRA-Vintage |
| Pop Art | strangerzonehf/Wan2.1-LoRA-Pop-Art |
| Minimalist | prithivMLmods/Wan-Minimalist-LoRA |
| Horror | prithivMLmods/Wan-Horror-LoRA |
| Concept Art | prithivMLmods/Wan-Concept-Art-LoRA |

### Hardware Requirements

- **Minimum**: CPU (slower generation)
- **Recommended**: GPU with 8GB+ VRAM for fast generation
- **Optimal**: GPU with 12GB+ VRAM for larger resolutions

## 📦 Installation (Local)

To run this Space locally:

```bash
# Clone the repository
git clone https://huggingface.co/spaces/YOUR_USERNAME/magic-wan-image-v2
cd magic-wan-image-v2

# Install dependencies
pip install -r requirements.txt

# Run the app
python app.py
```

## 🔗 Links & Resources

- **Base Model**: [Magic-Wan-Image-V2](https://huggingface.co/wikeeyang/Magic-Wan-Image-V2)
- **Diffusers Documentation**: [Hugging Face Diffusers](https://huggingface.co/docs/diffusers)
- **Gradio Documentation**: [Gradio 6 Docs](https://gradio.app/docs)
- **Built with**: [anyCoder](https://huggingface.co/spaces/akhaliq/anycoder)

## 📄 License

This Space is licensed under the [Apache 2.0 License](https://www.apache.org/licenses/LICENSE-2.0).

Please check the individual LoRA model licenses for their specific terms.

## 🙏 Credits

- **Base Model**: [@wikeeyang](https://huggingface.co/wikeeyang) for Magic-Wan-Image-V2
- **LoRA Creators**: Various creators on Hugging Face Hub
- **Framework**: [Gradio](https://gradio.app) team
- **Built with**: [anyCoder](https://huggingface.co/spaces/akhaliq/anycoder)

## 🐛 Issues & Support

If you encounter any issues:

1. Check the console logs for error messages
2. Try reducing image resolution
3. Ensure your prompt is not empty
4. Report issues on the Space discussion tab

---

<div align="center">

**Made with ❤️ using Gradio 6**

[Built with anycoder](https://huggingface.co/spaces/akhaliq/anycoder) | [Report Issue](https://huggingface.co/spaces/YOUR_USERNAME/magic-wan-image-v2/discussions)

</div>
```