Final version: Better hands, faces, eyes. Reduced NSFW content. LCM with simple, normal, or beta settings. Steps: 6–12. CFG: 1–1.4. Clip skip 2. Can generate at 832×1216 or 1024×1536.My custom-merged model produces exceptional results when combined with [LoRAs]. Give it a try!<h3 id="v1-4-pairs-exceptionally-well-with-simulacrum-clip-l-and-clip-g.">V1-4 pairs exceptionally well with Simulacrum <a target="_blank" rel="ugc" href="https://civarchive.com/models/1044804?modelVersionId=1208216">CLIP-L and CLIP-G.</a></h3><h2 id="v3-prompt-tips">🖌️ V3 Prompt Tips</h2><pre><code># POSITIVE PROMPT STARTER:
"masterpiece, Best quality, [your description here]"

# NEGATIVE PROMPT STARTER:
"low quality, worst quality, [unwanted elements]"</code></pre>## 🌟 Version Recommendations- V2: Best for photorealistic outputs- V1: more like V2 than V3, still great for person loras- V3: Optimized for artistic quality## ⚙️ Recommended Settings (V1-V2)```yamlSamplers:- DPM++ 2M- DPM++ 2M SDESampling Steps: 30-50Clip Skip: 1-2CFG Scale: 4-6<h2 id="optional-enhancements">🔧 Optional Enhancements</h2><ul><li>Use ADetailer for refined faces/hands</li><li>Enable Hires.fix for upscaling (recommended: 1.5-2x scale)</li></ul>

NaturalBeautyXL

better detail, better overall, but the same

naturalbeautyxl_finalFP32.safetensors

Final_FP32

Final

Retro Movies

v4_NSFW

v3.0

NaturalBeauty_inpaintv1.0

v2.0

v1.0

Final version: Better hands, faces, eyes. Reduced NSFW content. LCM with simple, normal, or beta settings. Steps: 6–12. CFG: 1–1.4. Clip skip 2. Can generate at 832×1216 or 1024×1536.

naturalbeautyxl_final.safetensors

Final version: Better hands, faces, eyes. Reduced NSFW content. LCM with simple, normal, or beta settings. Steps: 6–12. CFG: 1–1.4. Clip skip 2. Can generate at 832×1216 or 1024×1536.My custom-merged model produces exceptional results when combined with [LoRAs]. Give it a try!<h3 id="v1-4-pairs-exceptionally-well-with-simulacrum-clip-l-and-clip-g.">V1-4 pairs exceptionally well with Simulacrum <a target="_blank" rel="ugc" href="https://civitai.com/models/1044804?modelVersionId=1208216">CLIP-L and CLIP-G.</a></h3><h2 id="v3-prompt-tips">🖌️ V3 Prompt Tips</h2><pre><code># POSITIVE PROMPT STARTER:
"masterpiece, Best quality, [your description here]"

# NEGATIVE PROMPT STARTER:
"low quality, worst quality, [unwanted elements]"</code></pre>## 🌟 Version Recommendations- V2: Best for photorealistic outputs- V1: more like V2 than V3, still great for person loras- V3: Optimized for artistic quality## ⚙️ Recommended Settings (V1-V2)```yamlSamplers:- DPM++ 2M- DPM++ 2M SDESampling Steps: 30-50Clip Skip: 1-2CFG Scale: 4-6<h2 id="optional-enhancements">🔧 Optional Enhancements</h2><ul><li>Use ADetailer for refined faces/hands</li><li>Enable Hires.fix for upscaling (recommended: 1.5-2x scale)</li></ul>

Files with SHA256: c36d4cdb917290dea05a0483b9ec2367ef663ae3dfa87700df7291ff20023d71