diff --git a/.env.example b/.env.example
index 278923c..8b373af 100644
--- a/.env.example
+++ b/.env.example
@@ -15,10 +15,33 @@ HF_TOKEN=hf_your_token_here
 # Get it with: tailscale ip -4
 # GPU_TAILSCALE_IP=100.100.108.13
 
+# ============================================================================
+# CivitAI API Key (Optional, for NSFW models)
+# ============================================================================
+# Get your API key from: https://civitai.com/user/account
+# Required for downloading NSFW models with artifact_civitai_download.sh
+# CIVITAI_API_KEY=your_civitai_api_key_here
+
+# ============================================================================
+# Tailscale VPN (Optional, for remote access)
+# ============================================================================
+# Get your authkey from: https://login.tailscale.com/admin/settings/keys
+# TAILSCALE_AUTHKEY=tskey-auth-your_tailscale_authkey_here
+
+# ============================================================================
+# WebDAV Sync (Optional, for output synchronization)
+# ============================================================================
+# WebDAV credentials for syncing ComfyUI outputs to cloud storage
+# WEBDAV_URL=https://webdav.example.com/
+# WEBDAV_USERNAME=your_username
+# WEBDAV_PASSWORD=your_password
+# WEBDAV_REMOTE_PATH=/path/to/comfyui/outputs
+
 # ============================================================================
 # Notes
 # ============================================================================
 # - HF_TOKEN is the only required variable for basic operation
+# - CIVITAI_API_KEY is required for downloading NSFW models from CivitAI
 # - Models will be cached in /workspace/ directories on RunPod
 # - Orchestrator automatically manages model switching
 # - No database credentials needed (stateless architecture)
diff --git a/CLAUDE.md b/CLAUDE.md
index 6b18e39..bf58c6f 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -288,6 +288,79 @@ environment=HF_HOME="../huggingface_cache",HF_TOKEN="%(ENV_HF_TOKEN)s"
 supervisorctl start vllm-mistral
 ```
 
+### Downloading NSFW Models from CivitAI
+
+The project includes dedicated tooling for downloading NSFW models from CivitAI, separate from the HuggingFace downloader.
+
+**Files:**
+- Configuration: `models_civitai.yaml` - Defines available CivitAI models
+- Downloader: `artifact_civitai_download.sh` - Beautiful CLI tool for downloading models
+
+**Requirements:**
+- CivitAI API key (set in `.env` file as `CIVITAI_API_KEY`)
+- Get your key from: https://civitai.com/user/account
+
+**Available Models (Essential):**
+1. **LUSTIFY v7.0 GGWP** (6.31GB) - Photoreal NSFW checkpoint for women
+2. **Pony Diffusion V6 XL** (6.5GB) - Anime/furry NSFW with balanced content
+3. **RealVisXL V5.0** (6.8GB) - Photorealistic NSFW model
+
+**Usage:**
+
+```bash
+# Download all essential NSFW models (19.61GB)
+artifact_civitai_download.sh
+
+# Download only (no symlinks)
+artifact_civitai_download.sh download
+
+# Create symlinks only (models already downloaded)
+artifact_civitai_download.sh link
+
+# Custom configuration file
+artifact_civitai_download.sh -c /path/to/models_civitai.yaml
+
+# Help
+artifact_civitai_download.sh --help
+```
+
+**Directory Structure:**
+- Models downloaded to: `/workspace/models/civitai/` (cache)
+- Symlinks created in: `/workspace/ComfyUI/models/checkpoints/`
+
+**Features:**
+- Beautiful purple/magenta themed CLI (vs cyan for HuggingFace)
+- Automatic retry with exponential backoff for rate limiting
+- Resume interrupted downloads
+- Progress bars and status indicators
+- Parallel download support (respects rate limits)
+
+**Configuration Example (`models_civitai.yaml`):**
+
+```yaml
+model_categories:
+  nsfw_checkpoints:
+    - name: lustify-v7-ggwp
+      version_id: "1094291"
+      model_id: "573152"
+      description: "LUSTIFY v7.0 GGWP - Photoreal NSFW checkpoint"
+      size_gb: 6.31
+      essential: true
+      type: checkpoints
+```
+
+**Installation Profiles:**
+- `essential`: 3 models (LUSTIFY, Pony, RealVisXL) - 19.61GB, 30-60 min
+- `complete`: All 6 models - 39.21GB, 1-2 hours
+- `realistic_only`: LUSTIFY + RealVisXL - 13.11GB, 20-40 min
+- `anime_only`: Pony + WAI-NSFW - 13.0GB, 20-40 min
+
+**Important Notes:**
+- All models are SDXL-based (require ~12GB VRAM)
+- Cannot run simultaneously with FLUX models (both use ~23GB)
+- Can run with smaller models like standard SDXL (~12GB total)
+- Rate limiting: 5 second delay between downloads (configurable)
+
 ## Key Implementation Details
 
 ### Direct vLLM Server Architecture
diff --git a/models_civitai.yaml b/models_civitai.yaml
new file mode 100644
index 0000000..fd88b1c
--- /dev/null
+++ b/models_civitai.yaml
@@ -0,0 +1,206 @@
+# ============================================================================
+# CivitAI Model Configuration
+# ============================================================================
+#
+# This configuration file defines all available CivitAI models for download.
+# Models are organized by category: NSFW checkpoints, LoRAs, embeddings, etc.
+#
+# Each model entry contains:
+#   - name: Filesystem-friendly model name
+#   - version_id: CivitAI version ID (used for downloads)
+#   - model_id: CivitAI model ID (used for metadata queries)
+#   - description: Human-readable description
+#   - size_gb: Approximate size in gigabytes
+#   - essential: Whether this is an essential model (true/false)
+#   - type: Model type (checkpoints, loras, embeddings, etc.)
+#   - tags: Array of descriptive tags
+#
+# ============================================================================
+
+# Global settings
+settings:
+  cache_dir: /workspace/models/civitai
+  output_dir: /workspace/ComfyUI/models
+  parallel_downloads: 1
+  retry_attempts: 3
+  timeout_seconds: 3600
+  rate_limit_delay: 5  # seconds between downloads
+
+# Model categories
+model_categories:
+  # ==========================================================================
+  # NSFW IMAGE GENERATION MODELS
+  # ==========================================================================
+  nsfw_checkpoints:
+    - name: lustify-v7-ggwp
+      version_id: "1094291"
+      model_id: "573152"
+      description: "LUSTIFY v7.0 GGWP - Photoreal NSFW checkpoint for women in sexual scenarios"
+      size_gb: 6.31
+      essential: true
+      type: checkpoints
+      format: safetensors
+      base_model: SDXL 1.0
+      vram_gb: 12
+      tags: [nsfw, realistic, photography, women]
+      recommended_settings:
+        sampler: "DPM++ 2M SDE / DPM++ 3M SDE"
+        scheduler: "Exponential / Karras"
+        steps: 30
+        cfg_scale: "4-7"
+      notes: "Understands both danbooru tags and natural language prompting"
+
+    - name: pony-diffusion-v6-xl
+      version_id: "135867"
+      model_id: "257749"
+      description: "Pony Diffusion V6 XL - Versatile anime/cartoon/furry model with balanced content"
+      size_gb: 6.5
+      essential: true
+      type: checkpoints
+      format: safetensors
+      base_model: SDXL 1.0
+      vram_gb: 12
+      tags: [nsfw, anime, furry, cartoon, versatile]
+      training_info:
+        images: "2.6M aesthetically ranked"
+        ratio: "1:1:1 safe/questionable/explicit"
+      notes: "Most popular anime/furry model on CivitAI, supports wide range of styles"
+
+    - name: realvisxl-v5
+      version_id: "904692"
+      model_id: "139562"
+      description: "RealVisXL V5.0 - Photorealistic model for high-quality realistic images"
+      size_gb: 6.8
+      essential: true
+      type: checkpoints
+      format: safetensors
+      base_model: SDXL 1.0
+      vram_gb: 12
+      tags: [nsfw, realistic, photorealistic, professional]
+      recommended_settings:
+        sampler: "DPM++ SDE Karras (30+ steps) or DPM++ 2M Karras (50+ steps)"
+      notes: "Excellent for photorealistic portraits and scenes, both SFW and NSFW"
+
+    - name: wai-nsfw-illustrious-sdxl
+      version_id: "1239648"
+      model_id: "827184"
+      description: "WAI-NSFW-illustrious-SDXL v15.0 - Actively updated NSFW Illustrious variant"
+      size_gb: 6.5
+      essential: false
+      type: checkpoints
+      format: safetensors
+      base_model: SDXL 1.0
+      vram_gb: 12
+      tags: [nsfw, anime, illustrious, artistic]
+      notes: "Latest version from October 2025, frequently updated"
+
+    - name: talmendoxl
+      version_id: "131869"
+      model_id: "119202"
+      description: "TalmendoXL - Uncensored SDXL model biased towards photorealism"
+      size_gb: 6.6
+      essential: false
+      type: checkpoints
+      format: safetensors
+      base_model: SDXL 1.0
+      vram_gb: 12
+      tags: [nsfw, realistic, uncensored, photorealistic]
+      notes: "Known for high-quality photorealistic NSFW content"
+
+    - name: big-lust-v1-6
+      version_id: "649023"
+      model_id: "575395"
+      description: "Big Lust v1.6 - SDXL NSFW checkpoint"
+      size_gb: 6.5
+      essential: false
+      type: checkpoints
+      format: safetensors
+      base_model: SDXL 1.0
+      vram_gb: 12
+      tags: [nsfw, versatile]
+      notes: "General-purpose NSFW model with good results across styles"
+
+# ============================================================================
+# STORAGE & VRAM SUMMARIES
+# ============================================================================
+
+storage_requirements:
+  essential_only:
+    nsfw_checkpoints: 19.61  # LUSTIFY + Pony + RealVisXL
+    total: 19.61
+
+  all_models:
+    nsfw_checkpoints: 39.21  # All 6 checkpoints
+    total: 39.21
+
+vram_requirements:
+  # For 24GB GPU (RTX 4090)
+  notes: |
+    All SDXL checkpoints require approximately 12GB VRAM.
+    Can run one checkpoint at a time on 24GB GPU.
+    For simultaneous loading with other models (ComfyUI, vLLM),
+    ensure combined VRAM usage stays under 24GB.
+
+# ============================================================================
+# INSTALLATION PROFILES
+# ============================================================================
+
+installation_profiles:
+  essential:
+    description: Essential NSFW models only (3 checkpoints)
+    categories: [nsfw_checkpoints]
+    essential_only: true
+    storage_gb: 19.61
+    estimated_time: "30-60 minutes"
+    models:
+      - lustify-v7-ggwp
+      - pony-diffusion-v6-xl
+      - realvisxl-v5
+
+  complete:
+    description: All NSFW models (6 checkpoints)
+    categories: [nsfw_checkpoints]
+    storage_gb: 39.21
+    estimated_time: "1-2 hours"
+    models:
+      - lustify-v7-ggwp
+      - pony-diffusion-v6-xl
+      - realvisxl-v5
+      - wai-nsfw-illustrious-sdxl
+      - talmendoxl
+      - big-lust-v1-6
+
+  realistic_only:
+    description: Photorealistic NSFW models only
+    categories: [nsfw_checkpoints]
+    storage_gb: 13.11
+    estimated_time: "20-40 minutes"
+    models:
+      - lustify-v7-ggwp
+      - realvisxl-v5
+
+  anime_only:
+    description: Anime/cartoon NSFW models only
+    categories: [nsfw_checkpoints]
+    storage_gb: 13.0
+    estimated_time: "20-40 minutes"
+    models:
+      - pony-diffusion-v6-xl
+      - wai-nsfw-illustrious-sdxl
+
+# ============================================================================
+# METADATA
+# ============================================================================
+
+metadata:
+  version: 1.0.0
+  last_updated: 2025-11-23
+  compatible_with:
+    - ComfyUI >= 0.1.0
+    - SDXL base models
+  maintainer: Valknar
+  repository: https://github.com/valknar/runpod
+  notes: |
+    All models listed are NSFW (adult content) and hosted on CivitAI.
+    Requires CIVITAI_API_KEY for downloads.
+    Models will be downloaded to cache_dir and symlinked to ComfyUI.
diff --git a/comfyui_models.yaml b/models_huggingface.yaml
similarity index 100%
rename from comfyui_models.yaml
rename to models_huggingface.yaml