Commit
•
10230fd
1
Parent(s):
23c17b2
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,144 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
tags:
|
| 5 |
+
- art
|
| 6 |
+
- image-generation
|
| 7 |
+
- text-to-image
|
| 8 |
+
- diffusion
|
| 9 |
+
- high-quality
|
| 10 |
+
- AI
|
| 11 |
+
pipeline_tag: text-to-image
|
| 12 |
+
thumbnail: https://example.com/thumbnail.png
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# AnimeSAI V0.1
|
| 16 |
+
|
| 17 |
+
**Model Card for `enhanceaiteam/AnimeSAI`**
|
| 18 |
+
|
| 19 |
+
`AnimeSAI` is a cutting-edge text-to-image generation model by EnhanceAI, tailored specifically for creating high-quality anime-style images. Utilizing advanced diffusion techniques, this model is ideal for artists, designers, and anime enthusiasts looking to produce visually stunning and accurate anime art based on textual descriptions.
|
| 20 |
+
|
| 21 |
+
## Model Details
|
| 22 |
+
|
| 23 |
+
- **Version**: 0.1
|
| 24 |
+
- **Base Model**: EnhanceAI's proprietary image generation architecture
|
| 25 |
+
- **Training Data**: The model is trained on a diverse dataset of high-resolution, copyright-free anime images, covering various styles and themes.
|
| 26 |
+
- **Pipeline**: Stable Diffusion XL Pipeline
|
| 27 |
+
|
| 28 |
+
## Features
|
| 29 |
+
|
| 30 |
+
- **Anime-Specific Generation**: Specializes in generating anime-style art, from characters to landscapes.
|
| 31 |
+
- **High-Resolution Output**: Generates images at 1024x1024 resolution.
|
| 32 |
+
- **Enhanced Prompt Understanding**: Optimized for comprehending complex and detailed text prompts specific to anime art.
|
| 33 |
+
- **Versatile Style Generation**: Capable of producing images in a range of anime styles, from classic to modern aesthetics.
|
| 34 |
+
|
| 35 |
+
## Usage
|
| 36 |
+
|
| 37 |
+
To use `AnimeSAI`, you can integrate it with the `diffusers` library. Below is an example of how to generate images:
|
| 38 |
+
|
| 39 |
+
```python
|
| 40 |
+
import torch
|
| 41 |
+
from diffusers import (
|
| 42 |
+
StableDiffusionXLPipeline,
|
| 43 |
+
KDPM2AncestralDiscreteScheduler,
|
| 44 |
+
AutoencoderKL
|
| 45 |
+
)
|
| 46 |
+
|
| 47 |
+
# Load VAE component
|
| 48 |
+
vae = AutoencoderKL.from_pretrained(
|
| 49 |
+
"madebyollin/sdxl-vae-fp16-fix",
|
| 50 |
+
torch_dtype=torch.float16
|
| 51 |
+
)
|
| 52 |
+
|
| 53 |
+
# Configure the pipeline
|
| 54 |
+
pipe = StableDiffusionXLPipeline.from_pretrained(
|
| 55 |
+
"enhanceaiteam/AnimeSAI",
|
| 56 |
+
vae=vae,
|
| 57 |
+
torch_dtype=torch.float16
|
| 58 |
+
)
|
| 59 |
+
pipe.scheduler = KDPM2AncestralDiscreteScheduler.from_config(pipe.scheduler.config)
|
| 60 |
+
pipe.to('cuda')
|
| 61 |
+
|
| 62 |
+
# Define prompts and generate image
|
| 63 |
+
prompt = "a vibrant anime character standing under cherry blossoms"
|
| 64 |
+
negative_prompt = ""
|
| 65 |
+
|
| 66 |
+
image = pipe(
|
| 67 |
+
prompt,
|
| 68 |
+
negative_prompt=negative_prompt,
|
| 69 |
+
width=1024,
|
| 70 |
+
height=1024,
|
| 71 |
+
guidance_scale=7,
|
| 72 |
+
num_inference_steps=50,
|
| 73 |
+
clip_skip=2
|
| 74 |
+
).images[0]
|
| 75 |
+
|
| 76 |
+
image.save("generated_anime_image.png")
|
| 77 |
+
```
|
| 78 |
+
|
| 79 |
+
## Prompts
|
| 80 |
+
|
| 81 |
+
When creating prompts, be as descriptive as possible to achieve the best results. Detailed descriptions help the model generate more accurate and visually appealing anime images.
|
| 82 |
+
|
| 83 |
+
## Image Samples
|
| 84 |
+
|
| 85 |
+
Here are some example images generated by `AnimeSAI`:
|
| 86 |
+
|
| 87 |
+
1. 
|
| 88 |
+
2. 
|
| 89 |
+
3. 
|
| 90 |
+
4. 
|
| 91 |
+
5. 
|
| 92 |
+
6. 
|
| 93 |
+
7. 
|
| 94 |
+
8. 
|
| 95 |
+
9. 
|
| 96 |
+
10. 
|
| 97 |
+
11. 
|
| 98 |
+
13. 
|
| 99 |
+
14. 
|
| 100 |
+
15. 
|
| 101 |
+
16. 
|
| 102 |
+
17. 
|
| 103 |
+
18. 
|
| 104 |
+
19. 
|
| 105 |
+
20. 
|
| 106 |
+
21. 
|
| 107 |
+
22. 
|
| 108 |
+
23. 
|
| 109 |
+
24. 
|
| 110 |
+
25. 
|
| 111 |
+
|
| 112 |
+
## Explore
|
| 113 |
+
|
| 114 |
+
Explore and try out `AnimeSAI` on the [EnhanceAI Playground](https://enhanceai.art/playground). Experiment with different prompts and see the incredible anime-style images you can create.
|
| 115 |
+
|
| 116 |
+
## API Service
|
| 117 |
+
|
| 118 |
+
- **Modelslab** - [AnimeSAI on Modelslab](https://modelslab.com/models/animesai)
|
| 119 |
+
|
| 120 |
+
## License
|
| 121 |
+
|
| 122 |
+
`AnimeSAI` is licensed under the [Apache 2.0 License](https://opensource.org/licenses/Apache-2.0).
|
| 123 |
+
|
| 124 |
+
## Contact
|
| 125 |
+
|
| 126 |
+
For support and inquiries, please reach out to [info@enhanceai.art](mailto:info@enhanceai.art).
|
| 127 |
+
|
| 128 |
+
---
|
| 129 |
+
|
| 130 |
+
### How to Use AnimeSAI on EnhanceAI.art
|
| 131 |
+
|
| 132 |
+
1. **Open the Website**: Visit [EnhanceAI.art](https://enhanceai.art).
|
| 133 |
+
2. **Login/Register**: Create an account or log in.
|
| 134 |
+
3. **Access Playground V2**: Navigate to the Playground V2 section.
|
| 135 |
+
4. **Select Model**: Choose `AnimeSAI` from the available models.
|
| 136 |
+
5. **Write Prompt**: Enter your desired text prompt.
|
| 137 |
+
6. **Click Generate**: Hit generate and watch your anime-style image come to life.
|
| 138 |
+
|
| 139 |
+
---
|
| 140 |
+
|
| 141 |
+
Happy creating!
|
| 142 |
+
|
| 143 |
+
- The EnhanceAI Team
|
| 144 |
+
- Pranav Ajay
|