Xiaolxl Guofeng V1

🚀 Gf_style - Introduction

Welcome to the Gf_style model - a Chinese gorgeous ancient - style model, which can also be regarded as an ancient - style game character model with a 2.5D texture effect. This is a series of models that will be continuously updated in the future.

Version 2.0 has been released: https://huggingface.co/xiaolxl/Gf_style2

Version 3.0 has been released: https://huggingface.co/xiaolxl/Gf_style3

🚀 Quick Start

This model offers a unique Chinese ancient - style aesthetic, suitable for generating various ancient - style game character images or related artworks.

✨ Features

• Chinese Gorgeous Ancient Style: Presents a distinct Chinese ancient - style charm, with a 2.5D texture effect.
• Model Series: Continuously updated in the future to bring more features and improvements.

📦 Installation

1. Put the XXX.safetensors model and XXX.yaml into the SD directory.
2. Remember to select any VAE file, otherwise the drawing will be gray.

💻 Usage Examples

Basic Usage

(TIP: The characters are generated in vertical format. Theoretically, generating vertical images will yield better results.)

If you want to make the picture better as possible, please try the following configuration:

• Sampling steps: 30 or 50
• Sampler: DDIM or (DPM++ 2M Karras, DPM++ SDE Karras) - These two have different surprises.
• The size of the picture should be at least 768, otherwise it will collapse.
• Turn on Hires fix: R - ESRGAN 4x+ Anime6B and Upscale by 2
• If the face is deformed, try to Open face repair
• Key word(Start):

{best quality}, {{masterpiece}}, {highres}, {an extremely delicate and beautiful}, original, extremely detailed wallpaper,

• Negative words:

NSFW, lowres,bad anatomy,bad hands, text, error, missing fingers,extra digit, fewer digits, cropped, worstquality, low quality, normal quality,jpegartifacts,signature, watermark, username,blurry,bad feet

Advanced Usage

There are no specific advanced usage examples provided in the original text. You can further explore different parameter combinations and keyword settings according to your needs.

📚 Documentation

Examples

(You can find the original image in the file list and put it in WebUi to view keywords and other information)

Town building map

Graph generated by keywords in How to use

Ending figure

📄 License

The model is licensed under the creativeml - openrail - m license.

⚠️ Important Note

The characters are generated in vertical format. Theoretically, generating vertical images will yield better results. The size of the picture should be at least 768, otherwise it will collapse. Remember to select any VAE file, otherwise the drawing will be gray.

💡 Usage Tip

Try different sampler settings like DDIM or (DPM++ 2M Karras, DPM++ SDE Karras) as they may bring different surprises. If the face is deformed, try to open face repair.