basesssssp's picture
Update README.md
993a010 verified
metadata
license: openrail
language:
  - en
base_model:
  - Laxhar/noobai-XL-Vpred-1.0
pipeline_tag: text-to-image
tags:
  - art
  - NoobAI
  - SDXL
  - merge
  - cyberfix

noobaiXLNAIXL_vPred10Version-cyberrealisticv5-perpendicular

A merged Stable Diffusion XL model combining specialized fine-tuned components for enhanced anime-style generation and general coherence. Built using the same merging method as Doctor-Shotgun/NoobAI-XL-Merges.

📝 Model Details

  • Mechanism: Merge of three models:
    • NoobAI: Fine-tuned for anime/niche domains.
    • SDXL1.0 + 0.9VAE: Base model for general-purpose stability.
    • cyberrealisticv5: Enhances realism and detail coherence.
  • Result: NoobAIcyberfix—improves detail retention, human anatomy.

⚙️ Recommended Parameters

  • CFG Scale: 5 (default)
  • CFG-fix (optional): 4.5