# Emi Model Card

![eyecatch.png](eyecatch.png)

# Introduction
Emi (Ethereal master of illustration) is a text-to-image specialized in AI Art by AI Picasso Inc. using H100 and SDXL 1.0.
Emi is NOT made of unauthorized reproduction images on Danbooru.

# Licence
Emi's licence is CreativeML Open RAIL++-M License.
Therefore, you can use Emi for **commercial purpose**.

# Usage
[Click Here](https://huggingface.co/spaces/aipicasso/emi-latest-demo) to use Demo..

[Click Here](emi.safetensors) to get Emi.

[Click Here](emi.safetensors) to get Emi stable.

# Examples


![example_1.jpg](example_1.jpg)

```
positive prompt: anime artwork, anime style, (1girl), (black bob hair:1.5), brown eyes, red maples, sky, ((transparent))
negative prompt: (embedding:unaestheticXLv31:0.5), photo, deformed, realism, disfigured, low contrast, bad hand
```

![example_2.png](example_2.png)

```
positive prompt: monochrome, black and white, (japanese manga), mount fuji
negative prompt: (embedding:unaestheticXLv31:0.5), photo, deformed, realism, disfigured, low contrast, bad hand
```

![example_3.jpg](example_3.jpg)
```
positive prompt: (1man), focus, white wavy short hair, blue eyes, black shirt, white background, simple background
negative prompt: (embedding:unaestheticXLv31:0.5), photo, deformed, realism, disfigured, low contrast, bad hand
```

# Toward improving output
- Use "transparent" in your prompt.
- You can use the prompts of Waifu Diffusion and Stable Diffusion.
- Use [Textual Inversion](https://civitai.com/models/119032/unaestheticxl-or-negative-ti) in your negative prompt.
- You can use ChatGPT to improve your prompt.

# Legal information
We made Emi legally in Japan because Japanese government allows to use images for information analysis such as machine learning.

# Contact
Please feel free to send email: support@aipicasso.app