import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("uhuya/AsianVibesMix", torch_dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]AsianVibesMix (SDXL)
AsianVibesMix is a specialized SDXL-based model fine-tuned to achieve high-fidelity, photorealistic imagery with a focus on Asian aesthetics. It excels at rendering detailed skin textures, realistic lighting, and authentic facial features for portraiture and lifestyle photography.
Model Details
- Developed by: [uhuya]
- Model Type: Diffusion-based text-to-image generative model
- Base Model: SDXL v1.0
- Language(s): English (Prompts)
- License: MIT
- Main Focus: Photorealism, Asian portraits, Fashion, and Artistic expression.
Recommended Settings
For optimal results, the following parameters are suggested:
- Sampler: DPM++ 2M Karras or Euler a
- Steps: 30 - 45
- CFG Scale: 6 - 7.5
- Resolution: 1024x1024 (Square), 896x1152 (Portrait), or 1152x896 (Landscape)
- Clip Skip: 1 or 2
Prompting Guide
Positive Prompt:
(photorealistic:1.3), ultra-detailed, 8k, masterpiece, beautiful Asian woman, cinematic lighting, highly detailed skin pores, soft shadows, sharp focus, elegant attire.
Negative Prompt:
(worst quality, low quality:1.4), (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, watermark, text.
Limitations & Disclaimer
Users must adhere to the following:
- Responsible use of generative tools is required.
- Do not use this model to generate illegal content, non-consensual imagery, or any material that violates local or international laws.
- Please respect the Hugging Face Community Guidelines.
Show your support: If you enjoy this model, please leave a Like ❤️ on the repository!
- Downloads last month
- 87
Model tree for uhuya/AsianVibesMix
Base model
stabilityai/stable-diffusion-xl-base-1.0