Upload 2851499/3220121/README.md with huggingface_hub

2851499/3220121/README.md

@@ -0,0 +1,75 @@
+---
+license: other
+tags:
+- facial
+- style
+---
+
+# Facial Aftermath H3 - v1.0 - 3220121
+
+**Model Type**: LORA
+
+**Base Model**: MiniMax H3
+
+**Trigger Words**: None
+
+**Tags**: facial, style
+
+## Gallery
+
+<table>
+  <tr>
+    <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8e728d46-a8f6-40c9-9231-815a59a8ff76/original=true/139434950.mp4" width="200" controls muted autoplay loop></video></td>
+    <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e7a044ba-79a9-4045-83e9-b809631c5010/original=true/139434952.mp4" width="200" controls muted autoplay loop></video></td>
+  </tr>
+</table>
+
+## Description
+
+Gives woman a happy smile, faces covered with cum and a pleasure for men of culture.
+
+# How to Use This LoRA: Core Instructions
+
+## 🔹 What It Was Trained On
+
+- **Extreme Close-Up Framing:** Optimized for tight facial portraits where the face, neck, and upper chest dominate the frame. Shallow depth of field is structurally baked in, naturally softening backgrounds to isolate micro-details.
+
+- **Viscous Fluid Dynamics:** Specializes in rendering thick, glossy white fluid with realistic gravity-driven flow. Captures slow drips, pooling at the jawline/nose, and high-contrast specular highlights that emphasize weight, thickness, and texture over flat coverage.
+
+- **Makeup Degradation & Passive States:** Trained on the controlled breakdown of heavy cosmetics—smudged dark eyeshadow, melting mascara, and precisely one detached/askew fake eyelash paired with natural skin sheen (tears/sweat). Expressions lean toward hollow submission: half-lidded eyes, slightly parted lips, and subtle breath cues that convey exhaustion or quiet acceptance.
+
+- **Cinematic Lighting & Grain:** Cool, clinical key lighting against dark, out-of-focus backdrops. The model inherently applies a raw, high-grain photographic texture `3xtremerough`) to avoid over-polished or digital-looking results.
+
+## 🔹 MiniMax Prompt Structure
+
+Use this exact sequence for consistent generation. Token order matters: early elements establish composition and subject, while later tokens refine mood, texture, and style anchor.
+
+`[Subject + hair/outfit]` → `[Fluid placement & drip direction]` → `[Makeup degradation state]` → `[Expression & breathing cue]` → `[Lighting & background tone]` → `3xtremerough close-up photograph`
+
+*Example flow:*
+
+`Young woman with wet blonde hair in a pink crop top, thick white viscous fluid dripping from forehead to chin, smudged dark eyeshadow with one fake eyelash peeling off, tear-streaked glossy skin and lips parted in passive breath, cool clinical lighting against a blurred dark backdrop, 3xtremerough close-up photograph`
+
+## 🔹 Essential Trigger Keywords
+
+Group these strategically based on what you want to emphasize. Combine 1–2 per category for optimal parsing; avoid overloading.
+
+- **Fluid & Texture:** `thick white viscous fluid`, `glossy highlights`, `slow gravity drips`, `pooling at jawline/nose`, `heavy texture coating`
+
+- **Makeup Breakdown:** `smudged dark eyeshadow`, `melting mascara`, `one fake eyelash off/askew`, `tear-streaked skin`, `natural pore visibility`
+
+- **Expression & Mood:** `half-open eyes`, `lips parted`, `passive/submissive gaze`, `shallow breathing cue`, `quiet exhaustion`
+
+- **Style Anchor (Always include):** `3xtremerough close-up photograph`, `shallow depth of field`, `cool clinical lighting`, `raw high-grain finish`
+
+*Usage Tip:* Keep total prompt length under 60 words. Place fluid and makeup degradation tokens before expression cues to ensure the LoRA locks texture first, then applies mood.
+
+---
+
+Author: [barleduc](https://civitai.com/user/barleduc)
+
+Model: [CivitAI Model Page](https://civitai.com/models/2851499?modelVersionId=3220121)
+
+Archive: [CivArchive Page](https://civarchive.com/models/2851499?modelVersionId=3220121)
+
+<!-- Version: 20260812_upload -->