clandestine xl 1.0

Introduction

Clandestine XL 1.0 is a finetuned model from Kivotos XL 2.0, designed for generating sharp-lined anime illustrations. It builds upon popular AI-generated art styles, particularly those found on Pixiv, and aims to improve anatomy and hand renderings in stylized anime character generation.

Architecture

• Developer: Linaqruf
• Model Type: Stable Diffusion XL
• Base Model: Kivotos XL 2.0
• License: Fair AI Public License 1.0-SD

Training

The model is trained using a specific prompt structure to optimize the generation of high-quality images. It utilizes special tags inherited from Animagine XL 3.1, which enhance image quality, aesthetic appeal, and relevance.

Guide: Running Locally

1. Installation:
   Install the required libraries using the following command:

   pip install diffusers transformers accelerate safetensors --upgrade
   

2. Code Example:
   Use the provided Python script to generate images:

   import torch
   from diffusers import StableDiffusionXLPipeline
   
   pipe = StableDiffusionXLPipeline.from_pretrained(
       "yodayo-ai/clandestine-xl-1.0", 
       torch_dtype=torch.float16, 
       use_safetensors=True,
       custom_pipeline="lpw_stable_diffusion_xl",
       add_watermarker=False,
       variant="fp16"
   )
   pipe.to('cuda')
   
   prompt = "1girl, kazusa \(blue archive\), blue archive, solo, upper body, v, smile, looking at viewer, outdoors, night, masterpiece, best quality, very aesthetic, absurdres"
   negative_prompt = "nsfw, (low quality, worst quality:1.2), very displeasing, 3d, watermark, signature, ugly, poorly drawn"
   
   image = pipe(
       prompt, 
       negative_prompt=negative_prompt,
       width=832,
       height=1216, 
       guidance_scale=7,
       num_inference_steps=28
   ).images[0]
   
   image.save("./cat.png")
   

3. Recommended Hardware:
   Utilize cloud GPUs for optimal performance, such as those available through Google Cloud, AWS, or Azure.

License

Clandestine XL 1.0 is licensed under the Fair AI Public License 1.0-SD. Modifications must be shared under the same license with accessible source code, especially for network-accessible versions. Distribution must adhere to the same or similarly stringent licensing terms. More details can be found at freedevproject.org/faipl-1.0-sd.