Update README.md

README.md

@@ -29,38 +29,12 @@ Strong hair rendering (flow, highlights, and color fidelity)
 Consistent lighting response from soft natural light to harsh directional setups
 Good depth handling in both simple and complex environments
 Resistant to style hijacking (holds prompt intent well)
-🧪 Stress-Test Coverage
 
-This model has been validated against:
-
-Clothing edge cases (loose / layered garments)
-Hair complexity (long, flowing, multi-tone)
-Harsh and directional lighting scenarios
-Style drift / unintended painterly bias
-Environmental depth and perspective
-Expression variation and facial consistency
-Accessory stacking and detail density
-Non-female subject sanity checks
 ⚙️ Recommended Settings
 Resolution: 832×1216 (optimal balance of detail and stability)
-Sampler: DPM++ 2M Karras / similar
-Steps: 25–35
-CFG: 4.5 – 7 (sweet spot around ~5.5)
-Clip Skip: 1–2 depending on style preference
-🎯 Use Cases
-Character portraits
-Stylized realism / anime hybrids
-Fashion/editorial compositions
-Fantasy / sci-fi characters
-Animation pipelines (SDXL → WAN workflows)
-🎥 Workflow Compatibility
+Sampler: Res2 / Beta
+Steps: 50-75
+CFG: 6 - 10 (sweet spot around ~7.5)
+LORA Strength: 1-3
 
-Optimized for downstream use in:
 
-Image-to-Video pipelines (WAN / AnimateDiff-style workflows)
-Multi-phase generation setups
-High-consistency character sequences
-⚠️ Notes
-Tuned primarily for female character generation, but supports broader subjects
-Designed to avoid over-texturing artifacts seen in multi-pass workflows
-Works best with clear prompt structure and moderate CFG values