Sea Art Furry X L 1.0

Introduction

SeaArt-Furry-XL-1.0 is a model developed by SeaArt for generating high-quality furry art images. It's built on the Stable Diffusion XL framework and offers advanced capabilities in understanding and creating detailed furry imagery. The model is designed to enhance the artistic quality of furry art and support collaboration within the furry community.

Architecture

SeaArt-Furry-XL-1.0 is a diffusion-based text-to-image generative model that uses a Latent Diffusion Model architecture with two fixed, pre-trained text encoders: OpenCLIP-ViT/G and CLIP-ViT/L. It follows a similar structure to Stable Diffusion XL, enabling it to generate images from text prompts effectively.

Training

The training process involved analyzing millions of furry images to refine the model's ability to generate accurate and high-quality furry characters, including mammals and birds. The model incorporates extensive species calibration and artist style hints, allowing it to produce images that align with specific user preferences and styles.

Guide: Running Locally

Basic Steps

1. Install Required Libraries:

   pip install diffusers transformers accelerate safetensors --upgrade
   

2. Run Image Generation:

   import torch
   from diffusers import DiffusionPipeline
   
   pipe = DiffusionPipeline.from_pretrained(
       "SeaArtLab/SeaArt-Furry-XL-1.0", 
       torch_dtype=torch.float16, 
       use_safetensors=True, 
   )
   pipe.to('cuda')
   
   prompt = "canid, canine, fox, mammal, red_fox, true_fox, foxgirl83, photonoko, day, digitigrade, fluffy, fluffy_tail, fur, orange_body, orange_fur, orange_tail, solo, sunlight, tail, mid, 2018, digital_media_(artwork), hi_res, masterpiece"
   negative_prompt = "nsfw, lowres, (bad), text, error, fewer, extra, missing, worst quality, jpeg artifacts, low quality, watermark, unfinished, displeasing, oldest, early, chromatic aberration, signature, extra digits, artistic error, username, scan, [abstract]"
   
   image = pipe(
       prompt, 
       negative_prompt=negative_prompt,
       width=832,
       height=1216, 
       guidance_scale=7,
       num_inference_steps=28
   ).images[0]
   
   image.save("./output/seaart_test.png")
   

Suggested Cloud GPUs

For optimal performance, consider using cloud-based GPUs such as NVIDIA's A100 or V100 on platforms like AWS, Google Cloud, or Azure.

License

SeaArt-Furry-XL-1.0 is licensed under the Fair AI Public License 1.0-SD, which is compatible with the Stable Diffusion models' license. Key aspects include:

1. Modification Sharing: Changes to the model must be shared along with the original license.
2. Source Code Accessibility: If the modified version is network-accessible, provide a source code download link.
3. Distribution Terms: Must comply with this license or one with similar rules.
4. Compliance: Non-compliance issues must be resolved within 30 days, promoting transparency and adherence to open-source principles.

This license supports an open and collaborative environment, encouraging ethical development and use of the model.