initial

2026-03-27 18:52:05 +01:00 · 2025-06-28 21:44:26 -04:00 · 2025-06-28 21:44:26 -04:00 · 3d92e6f72e
commit 3d92e6f72e
12 changed files with 687 additions and 0 deletions
--- a/.gitignore
+++ b/.gitignore
@ -0,0 +1,6 @@
+.vscode
+ComfyUI
+custom_nodes
+input
+output
+models
--- a/README.md
+++ b/README.md
@ -0,0 +1,182 @@
+# ComfyUI ROCm Docker Image
+
+🔥 **ComfyUI with AMD ROCm support** - Run ComfyUI on AMD GPUs with optimized ROCm-compatible dependencies.
+
+[![Docker Pulls](https://img.shields.io/docker/pulls/corundex/comfyui-rocm)](https://hub.docker.com/r/corundex/comfyui-rocm) [![ROCm](https://img.shields.io/badge/ROCm-6.4+-green)](https://rocm.docs.amd.com/) [![AMD GPU](https://img.shields.io/badge/AMD-RX%206000%2B-red)](https://www.amd.com/en/graphics/radeon-rx-graphics)
+
+![ComfyUI Interface](Screenshot.png)
+*ComfyUI running on AMD ROCm with sample workflow and generated landscape image*
+
+## 📋 Version Information
+
+- **Base Image**: `rocm/pytorch:rocm6.4.1_ubuntu24.04_py3.12_pytorch_release_2.6.0`
+- **Python**: 3.12.10
+- **PyTorch**: 2.6.0+git684f6f2
+- **ROCm**: 6.4.43483-a187df25c
+- **ComfyUI**: v0.3.43 (e18f53c, 2025-06-27)
+
+## ✨ Key Features
+
+- 🎨 **Node-based AI workflow** - Visual interface for creating complex AI pipelines
+- 🔥 **AMD ROCm optimized** - Native AMD GPU acceleration with ROCm 6.4+
+- 📦 **Smart model management** - Automatic downloads with configurable model sets
+- 🧪 **Tested compatibility** - All dependencies verified on real AMD hardware
+- 🎯 **Ready to use** - Pre-configured with sample workflows
+- 💾 **Persistent storage** - Models and outputs preserved across restarts
+
+
+## 🚀 Quick Start
+
+```bash
+# Pull and run ComfyUI with ROCm support
+docker run -d \
+  --device=/dev/kfd \
+  --device=/dev/dri \
+  --group-add=video \
+  -p 8188:8188 \
+  -v $(pwd)/models:/workspace/ComfyUI/models \
+  -v $(pwd)/output:/workspace/ComfyUI/output \
+  corundex/comfyui-rocm:latest
+```
+
+Access ComfyUI at: **http://localhost:8188**
+
+## 📋 Requirements
+
+| Component | Requirement |
+|-----------|-------------|
+| **GPU** | AMD RX 6000/7000+ series with ROCm support |
+| **VRAM** | 8GB minimum (16GB+ recommended) |
+| **OS** | Linux (Ubuntu 24.04+ recommended) |
+| **Docker** | Latest version with GPU support |
+| **ROCm** | Drivers 6.4+ installed on host |
+
+## 🔧 Setup Instructions
+
+### 1. Install ROCm Drivers
+```bash
+# Ubuntu/Debian
+curl -fsSL https://repo.radeon.com/rocm/rocm.gpg.key | sudo gpg --dearmor -o /etc/apt/keyrings/rocm.gpg
+echo "deb [arch=amd64 signed-by=/etc/apt/keyrings/rocm.gpg] https://repo.radeon.com/rocm/apt/6.4 jammy main" | sudo tee /etc/apt/sources.list.d/rocm.list
+sudo apt update && sudo apt install rocm-dkms
+sudo usermod -a -G render,video $USER
+```
+
+### 2. Verify ROCm Installation
+```bash
+rocm-smi  # Should show your AMD GPU(s)
+```
+
+### 3. Run ComfyUI
+```bash
+docker run -d \
+  --name comfyui-rocm \
+  --device=/dev/kfd --device=/dev/dri --group-add=video \
+  -p 8188:8188 \
+  -v ./models:/workspace/ComfyUI/models \
+  -v ./output:/workspace/ComfyUI/output \
+  corundex/comfyui-rocm:latest
+```
+
+## 🎛️ Model Management
+
+Control model downloading with the `MODEL_DOWNLOAD` environment variable:
+
+| Mode | Description | Models Included |
+|------|-------------|-----------------|
+| `default` | Essential starter (4GB) | SD 1.5 |
+| `common` | Comprehensive set (~30GB) | SD 1.5, SDXL, ControlNets, upscalers, VAE, embeddings |
+| `realistic` | Photo-realistic models (~8GB) | Realistic Vision, DreamShaper, VAE |
+| `photorealistic` | SDXL realistic (~12GB) | Juggernaut XL, RealVisXL |
+| `artistic` | Creative/stylized (~2GB) | Deliberate v2 |
+| `all` | Everything (~100GB) | All model sets combined |
+| `none` | Skip downloads | Use existing models only |
+
+### Usage Examples
+
+```bash
+# Default models (SD 1.5)
+docker run -d --device=/dev/kfd --device=/dev/dri --group-add=video \
+  -p 8188:8188 -v ./models:/workspace/ComfyUI/models \
+  corundex/comfyui-rocm:latest
+
+# All models (~100GB download)
+docker run -d --device=/dev/kfd --device=/dev/dri --group-add=video \
+  -p 8188:8188 -e MODEL_DOWNLOAD=all \
+  -v ./models:/workspace/ComfyUI/models \
+  corundex/comfyui-rocm:latest
+
+# Use existing models only
+docker run -d --device=/dev/kfd --device=/dev/dri --group-add=video \
+  -p 8188:8188 -e MODEL_DOWNLOAD=none \
+  -v ./models:/workspace/ComfyUI/models \
+  corundex/comfyui-rocm:latest
+```
+
+## 🐳 Docker Compose
+
+```yaml
+services:
+  comfyui-rocm:
+    image: corundex/comfyui-rocm:latest
+    container_name: comfyui-rocm
+    devices:
+      - /dev/kfd:/dev/kfd
+      - /dev/dri:/dev/dri
+    group_add:
+      - video
+    ports:
+      - "8188:8188"
+    volumes:
+      - ./data/models:/workspace/ComfyUI/models
+      - ./data/output:/workspace/ComfyUI/output
+      - ./data/input:/workspace/ComfyUI/input
+      - ./data/custom_nodes:/workspace/ComfyUI/custom_nodes
+      - ./data/user:/workspace/ComfyUI/user
+    environment:
+      - MODEL_DOWNLOAD=default
+      - HIP_VISIBLE_DEVICES=0
+      - CUDA_VISIBLE_DEVICES=""
+    restart: unless-stopped
+```
+
+Run with: `docker compose up -d`
+
+## ⚡ Performance & Hardware
+
+### Tested Hardware
+- **AMD Radeon RX 9060 XT** (16GB VRAM) ✅
+
+### Performance Metrics
+- **Generation Time**: ~30-60s for 512x512 images
+- **VRAM Usage**: 4-8GB for basic operations  
+- **Model Loading**: ~30-60s first time, cached afterward
+- **Batch Processing**: Multiple images supported
+
+### Tips
+- Mount persistent volumes to avoid re-downloading models
+- Start with `default` models, upgrade to larger sets as needed
+- Use fast SSD storage for optimal performance
+
+## 🔍 Troubleshooting
+
+| Issue | Solution |
+|-------|----------|
+| **Container won't start** | Check ROCm drivers: `rocm-smi` |
+| **No GPU detected** | Verify container GPU access: `docker exec comfyui-rocm python -c "import torch; print(torch.cuda.is_available())"` |
+| **Model download fails** | Check internet connection, disk space, and logs |
+| **Out of memory** | Reduce batch size, use smaller models, ensure 8GB+ VRAM |
+| **Models not found** | Verify downloads completed and file permissions |
+
+## 📄 License & Credits
+
+This Docker image packages ComfyUI with ROCm support. ComfyUI is licensed under GPL-3.0.
+
+**Acknowledgments:**
+- [ComfyUI](https://github.com/comfyanonymous/ComfyUI) - Node-based AI workflow interface
+- [AMD ROCm](https://rocm.docs.amd.com/) - Open source GPU computing platform  
+- ROCm community for AMD GPU AI support
+
+---
+
+🔗 **Links:** [Docker Hub](https://hub.docker.com/r/corundex/comfyui-rocm) | [GitHub](https://github.com/corundex/comfyui_rocm) | [ComfyUI](https://github.com/comfyanonymous/ComfyUI)
--- a/Screenshot.png
+++ b/Screenshot.png
--- a/build.sh
+++ b/build.sh
@ -0,0 +1,17 @@
+#!/bin/bash
+# build.sh - Build ComfyUI ROCm Docker image
+
+set -e
+
+# Configuration
+IMAGE_NAME="comfyui-rocm"
+VERSION="${1:-latest}"
+DETAILED_TAG="comfyui_0.3.43__rocm6.4.1_ubuntu24.04_py3.12_pytorch_2.6.0"
+
+echo "🐳 Building ComfyUI ROCm Docker image..."
+echo "📦 Tags: ${IMAGE_NAME}:${VERSION}, ${IMAGE_NAME}:${DETAILED_TAG}"
+echo ""
+
+# Build the image with both tags
+echo "🔨 Building Docker image..."
+docker build -f docker/Dockerfile -t "${IMAGE_NAME}:${VERSION}" -t "${IMAGE_NAME}:${DETAILED_TAG}" --progress=plain .
--- a/docker-compose.yaml
+++ b/docker-compose.yaml
@ -0,0 +1,53 @@
+services:
+  comfyui-rocm:
+    image: corundex/comfyui-rocm:latest
+    pull_policy: always
+    container_name: comfyui-rocm
+    hostname: comfyui-rocm
+    
+    # GPU access for AMD ROCm
+    devices:
+      - /dev/kfd:/dev/kfd
+      - /dev/dri:/dev/dri
+    
+    # Add video group for GPU access
+    group_add:
+      - video
+    
+    # Port mapping
+    ports:
+      - "8188:8188"
+
+    # Volume mapping to preserve models and data
+    volumes:
+      # Model storage (checkpoints, VAE, LoRA, etc.)
+      - ./data/models:/workspace/ComfyUI/models
+      # Generated images and outputs
+      - ./data/output:/workspace/ComfyUI/output
+      # Input images for processing
+      - ./data/input:/workspace/ComfyUI/input
+      # Custom nodes and extensions
+      - ./data/custom_nodes:/workspace/ComfyUI/custom_nodes
+      # ComfyUI user settings and workflows
+      - ./data/user:/workspace/ComfyUI/user
+      # Temporary files (optional)
+      - ./data/temp:/workspace/ComfyUI/temp
+
+    # Environment variables
+    environment:
+      # Model download behavior: default, common, realistic, photorealistic, artistic, all, none
+      - MODEL_DOWNLOAD=default
+      # ROCm environment
+      - HIP_VISIBLE_DEVICES=0
+      - CUDA_VISIBLE_DEVICES=""
+
+    # Restart policy
+    restart: unless-stopped
+    
+    # Healthcheck
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:8188/"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 120s
--- a/docker/.dockerignore
+++ b/docker/.dockerignore
@ -0,0 +1,3 @@
+# Git
+.git
+.gitignore
--- a/docker/Dockerfile
+++ b/docker/Dockerfile
@ -0,0 +1,68 @@
+# ROCm ComfyUI Dockerfile
+# Based on AMD's official ROCm PyTorch container
+
+FROM rocm/pytorch:rocm6.4.1_ubuntu24.04_py3.12_pytorch_release_2.6.0
+
+# Set working directory
+WORKDIR /workspace
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+  git \
+  wget \
+  curl \
+  && apt-get clean \
+  && rm -rf /var/lib/apt/lists/*
+
+# Clone ComfyUI
+RUN git clone https://github.com/comfyanonymous/ComfyUI.git
+
+# Set ComfyUI as working directory
+WORKDIR /workspace/ComfyUI
+
+# Create necessary directories with separate commands
+RUN mkdir -p /workspace/ComfyUI/models/checkpoints && \
+  mkdir -p /workspace/ComfyUI/models/vae && \
+  mkdir -p /workspace/ComfyUI/models/loras && \
+  mkdir -p /workspace/ComfyUI/models/embeddings && \
+  mkdir -p /workspace/ComfyUI/models/upscale_models && \
+  mkdir -p /workspace/ComfyUI/models/controlnet && \
+  mkdir -p /workspace/ComfyUI/output && \
+  mkdir -p /workspace/ComfyUI/input && \
+  mkdir -p /workspace/ComfyUI/custom_nodes
+
+# Copy ROCm-tested requirements files
+COPY docker/requirements_rocm.txt /workspace/
+
+# Install Python dependencies using ROCm-compatible requirements
+# These files have been tested on real AMD hardware and filter out packages that break ROCm
+RUN pip install --no-cache-dir -r /workspace/requirements_rocm.txt
+
+# Copy startup script, models config, and sample workflow
+COPY docker/startup.sh /workspace/startup.sh
+COPY docker/download_models.py /workspace/download_models.py
+COPY docker/models.yaml /workspace/models.yaml
+COPY docker/sample_workflow.json /workspace/ComfyUI/
+
+# Make startup script executable
+RUN chmod +x /workspace/startup.sh
+
+# Environment variables for model download behavior
+# MODEL_DOWNLOAD options:
+#   "default"   - Download basic SD 1.5 model (default)
+#   "all"       - Download full model set (SD 1.5, SDXL, ControlNet, etc.)
+#   "realistic" - Download realistic photo models  
+#   "none"      - Skip all downloads, use existing models only
+#   Custom      - Any section name from models.conf
+ENV MODEL_DOWNLOAD=default
+
+# Expose ComfyUI port
+EXPOSE 8188
+
+# Health check to ensure ComfyUI is running
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+  CMD curl -f http://localhost:8188/ || exit 1
+
+# Use smart startup script that handles model downloads
+CMD ["/workspace/startup.sh"]
+
--- a/docker/download_models.py
+++ b/docker/download_models.py
@ -0,0 +1,131 @@
+#!/usr/bin/env python3
+"""
+Smart model downloader for ComfyUI
+"""
+import os
+import sys
+import yaml
+import requests
+from pathlib import Path
+from urllib.parse import urlparse
+
+def log(message):
+    print(f"[ComfyUI] {message}", flush=True)
+
+def download_file(url, filepath, name, min_size=0):
+    """Download file with progress bar and validation"""
+    if filepath.exists():
+        if filepath.stat().st_size >= min_size:
+            log(f"{name} already exists, skipping")
+            return True
+        else:
+            log(f"{name} exists but too small, redownloading")
+            filepath.unlink()
+    
+    log(f"Downloading {name}...")
+    log(f"URL: {url}")
+    log(f"Path: {filepath}")
+    
+    # Create directory if needed
+    filepath.parent.mkdir(parents=True, exist_ok=True)
+    
+    try:
+        response = requests.get(url, stream=True)
+        response.raise_for_status()
+        
+        with open(filepath, 'wb') as f:
+            for chunk in response.iter_content(chunk_size=8192):
+                if chunk:
+                    f.write(chunk)
+        
+        # Validate file size
+        if filepath.stat().st_size >= min_size:
+            log(f"{name} downloaded successfully")
+            return True
+        else:
+            log(f"{name} download failed - file too small")
+            filepath.unlink()
+            return False
+            
+    except Exception as e:
+        log(f"{name} download failed: {e}")
+        if filepath.exists():
+            filepath.unlink()
+        return False
+
+def load_models_config(config_path):
+    """Load models configuration from YAML"""
+    try:
+        with open(config_path, 'r') as f:
+            return yaml.safe_load(f)
+    except Exception as e:
+        log(f"Failed to load config {config_path}: {e}")
+        return None
+
+def download_model_set(models_config, model_set, models_base):
+    """Download a specific set of models"""
+    if model_set == 'all':
+        # Special case: download all model sets except 'all' itself
+        log("Downloading ALL model sets...")
+        all_success = True
+        for key in models_config['models']:
+            if key != 'all':  # Skip 'all' to avoid recursion
+                log(f"Processing model set: {key}")
+                if not download_model_set(models_config, key, models_base):
+                    all_success = False
+        log(f"All model sets completed: {'SUCCESS' if all_success else 'PARTIAL'}")
+        return all_success
+    
+    if model_set not in models_config['models']:
+        log(f"Unknown model set: {model_set}")
+        return False
+    
+    models = models_config['models'][model_set]
+    log(f"Downloading {model_set} models ({len(models)} total)...")
+    
+    success_count = 0
+    for model in models:
+        filepath = Path(models_base) / model['path']
+        min_size = model.get('min_size', 100000000)  # 100MB default
+        
+        if download_file(model['url'], filepath, model['name'], min_size):
+            success_count += 1
+    
+    log(f"{model_set} downloads completed: {success_count}/{len(models)} successful")
+    return success_count == len(models)
+
+def main():
+    models_base = "/workspace/ComfyUI/models"
+    config_path = "/workspace/models.yaml"
+    model_download = os.environ.get('MODEL_DOWNLOAD', 'default')
+    
+    log(f"Model downloader starting (MODEL_DOWNLOAD={model_download})")
+    
+    # Create model directories
+    for subdir in ['checkpoints', 'vae', 'loras', 'upscale_models', 'controlnet', 'embeddings']:
+        Path(models_base, subdir).mkdir(parents=True, exist_ok=True)
+    
+    if model_download == 'none':
+        log("Skipping model downloads (MODEL_DOWNLOAD=none)")
+        return True
+    
+    # Load configuration
+    models_config = load_models_config(config_path)
+    if not models_config:
+        log("Failed to load models configuration, skipping downloads")
+        return False
+    
+    # Download models
+    success = download_model_set(models_config, model_download, models_base)
+    
+    if success:
+        log("Model downloader completed successfully")
+    else:
+        log("Model downloader completed with some failures (continuing anyway)")
+    
+    # Always return success to avoid restart loops
+    return True
+
+if __name__ == '__main__':
+    success = main()
+    sys.exit(0 if success else 1)
--- a/docker/models.yaml
+++ b/docker/models.yaml
@ -0,0 +1,79 @@
+models:
+  default:
+    - name: "Stable Diffusion 1.5"
+      url: "https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.safetensors"
+      path: "checkpoints/v1-5-pruned-emaonly.safetensors"
+      min_size: 4000000000
+
+  common:
+    - name: "Stable Diffusion 1.5"
+      url: "https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.safetensors"
+      path: "checkpoints/v1-5-pruned-emaonly.safetensors"
+      min_size: 4000000000
+    - name: "SD 1.5 VAE"
+      url: "https://huggingface.co/stabilityai/sd-vae-ft-mse-original/resolve/main/vae-ft-mse-840000-ema-pruned.safetensors"
+      path: "vae/vae-ft-mse-840000-ema-pruned.safetensors"
+      min_size: 300000000
+    - name: "SDXL Base 1.0"
+      url: "https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors"
+      path: "checkpoints/sd_xl_base_1.0.safetensors"
+      min_size: 6000000000
+    - name: "SDXL Refiner 1.0"
+      url: "https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0/resolve/main/sd_xl_refiner_1.0.safetensors"
+      path: "checkpoints/sd_xl_refiner_1.0.safetensors"
+      min_size: 6000000000
+    - name: "ControlNet Canny"
+      url: "https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_canny.pth"
+      path: "controlnet/control_v11p_sd15_canny.pth"
+      min_size: 1400000000
+    - name: "ControlNet Depth"
+      url: "https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11f1p_sd15_depth.pth"
+      path: "controlnet/control_v11f1p_sd15_depth.pth"
+      min_size: 1400000000
+    - name: "ControlNet OpenPose"
+      url: "https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_openpose.pth"
+      path: "controlnet/control_v11p_sd15_openpose.pth"
+      min_size: 1400000000
+    - name: "Real-ESRGAN 4x Upscaler"
+      url: "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth"
+      path: "upscale_models/RealESRGAN_x4plus.pth"
+      min_size: 60000000
+    - name: "Real-ESRGAN 4x Anime"
+      url: "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth"
+      path: "upscale_models/RealESRGAN_x4plus_anime_6B.pth"
+      min_size: 15000000
+    - name: "EasyNegative Embedding"
+      url: "https://huggingface.co/datasets/gsdf/EasyNegative/resolve/main/EasyNegative.safetensors"
+      path: "embeddings/EasyNegative.safetensors"
+      min_size: 20000
+
+
+  realistic:
+    - name: "Realistic Vision v5.1"
+      url: "https://huggingface.co/SG161222/Realistic_Vision_V5.1_noVAE/resolve/main/Realistic_Vision_V5.1.safetensors"
+      path: "checkpoints/Realistic_Vision_V5.1.safetensors"
+      min_size: 2000000000
+    - name: "DreamShaper 8"
+      url: "https://huggingface.co/Lykon/DreamShaper/resolve/main/DreamShaper_8_pruned.safetensors"
+      path: "checkpoints/DreamShaper_8_pruned.safetensors"
+      min_size: 2000000000
+    - name: "Realistic VAE"
+      url: "https://huggingface.co/stabilityai/sd-vae-ft-mse-original/resolve/main/vae-ft-mse-840000-ema-pruned.safetensors"
+      path: "vae/vae-ft-mse-840000-ema-pruned.safetensors"
+      min_size: 300000000
+
+  photorealistic:
+    - name: "Juggernaut XL"
+      url: "https://huggingface.co/RunDiffusion/Juggernaut-XL-v9/resolve/main/Juggernaut-XL_v9_RunDiffusionPhoto_v2.safetensors"
+      path: "checkpoints/Juggernaut-XL_v9_RunDiffusionPhoto_v2.safetensors"
+      min_size: 6000000000
+    - name: "RealVisXL V4.0"
+      url: "https://huggingface.co/SG161222/RealVisXL_V4.0/resolve/main/RealVisXL_V4.0.safetensors"
+      path: "checkpoints/RealVisXL_V4.0.safetensors"
+      min_size: 6000000000
+
+  artistic:
+    - name: "Deliberate v2"
+      url: "https://huggingface.co/XpucT/Deliberate/resolve/main/Deliberate_v2.safetensors"
+      path: "checkpoints/Deliberate_v2.safetensors"
+      min_size: 2000000000
--- a/docker/requirements_rocm.txt
+++ b/docker/requirements_rocm.txt
@ -0,0 +1,53 @@
+# ROCm-compatible requirements.txt for ComfyUI
+# Generated on Fri Jun 27 23:19:36 EDT 2025
+# Base image: rocm/pytorch:latest
+# Each package tested in fresh ROCm container with 300s timeout
+#
+# Version Information:
+# - Python: 3.12.10
+# - PyTorch: 2.6.0+git684f6f2
+# - ROCm: 6.4.43483-a187df25c
+# - ComfyUI: e18f53c (2025-06-27)
+# - ComfyUI Tag: v0.3.43
+# - ComfyUI Version: 0.3.43
+# 
+# Legend:
+#   ✅ SAFE - can be installed without breaking ROCm
+#   ❌ BROKEN - breaks ROCm, fails to install, or times out
+#   📦 ALREADY - already installed correctly in base image
+
+comfyui-frontend-package==1.23.4
+comfyui-workflow-templates==0.1.30
+comfyui-embedded-docs==0.2.3
+# 📦 ALREADY: torch
+torchsde
+# 📦 ALREADY: torchvision
+# 🌙 NIGHTLY: torchaudio (use: pip install --pre torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm6.4)
+# 📦 ALREADY: numpy>=1.25.0
+einops
+transformers>=4.37.2
+tokenizers>=0.13.3
+sentencepiece
+safetensors>=0.4.2
+# 📦 ALREADY: aiohttp>=3.11.8
+# 📦 ALREADY: yarl>=1.18.0
+pyyaml
+Pillow
+# 📦 ALREADY: scipy
+# 📦 ALREADY: tqdm
+# 📦 ALREADY: psutil
+alembic
+SQLAlchemy
+kornia>=0.7.1
+spandrel
+# 📦 ALREADY: soundfile
+av>=14.2.0
+pydantic~=2.0
+pydantic-settings~=2.0
+
+# Testing Summary:
+# ✅ SAFE packages: 18
+# 📦 ALREADY installed: 9
+# 🌙 NIGHTLY packages: 1
+# ❌ BROKEN packages: 0
+# Total tested: 28
--- a/docker/sample_workflow.json
+++ b/docker/sample_workflow.json
@ -0,0 +1,80 @@
+{
+  "1": {
+    "inputs": {
+      "ckpt_name": "v1-5-pruned-emaonly.safetensors"
+    },
+    "class_type": "CheckpointLoaderSimple",
+    "_meta": {
+      "title": "Load Checkpoint"
+    }
+  },
+  "2": {
+    "inputs": {
+      "text": "a beautiful landscape with mountains and a lake, highly detailed, 8k, photorealistic",
+      "clip": ["1", 1]
+    },
+    "class_type": "CLIPTextEncode",
+    "_meta": {
+      "title": "CLIP Text Encode (Prompt)"
+    }
+  },
+  "3": {
+    "inputs": {
+      "text": "blurry, low quality, distorted, deformed, ugly, bad anatomy",
+      "clip": ["1", 1]
+    },
+    "class_type": "CLIPTextEncode",
+    "_meta": {
+      "title": "CLIP Text Encode (Negative)"
+    }
+  },
+  "4": {
+    "inputs": {
+      "width": 512,
+      "height": 512,
+      "batch_size": 1
+    },
+    "class_type": "EmptyLatentImage",
+    "_meta": {
+      "title": "Empty Latent Image"
+    }
+  },
+  "5": {
+    "inputs": {
+      "seed": 123456,
+      "steps": 20,
+      "cfg": 7,
+      "sampler_name": "euler_ancestral",
+      "scheduler": "normal",
+      "denoise": 1,
+      "model": ["1", 0],
+      "positive": ["2", 0],
+      "negative": ["3", 0],
+      "latent_image": ["4", 0]
+    },
+    "class_type": "KSampler",
+    "_meta": {
+      "title": "KSampler"
+    }
+  },
+  "6": {
+    "inputs": {
+      "samples": ["5", 0],
+      "vae": ["1", 2]
+    },
+    "class_type": "VAEDecode",
+    "_meta": {
+      "title": "VAE Decode"
+    }
+  },
+  "7": {
+    "inputs": {
+      "filename_prefix": "ComfyUI_AMD_test",
+      "images": ["6", 0]
+    },
+    "class_type": "SaveImage",
+    "_meta": {
+      "title": "Save Image"
+    }
+  }
+}
--- a/docker/startup.sh
+++ b/docker/startup.sh
@ -0,0 +1,15 @@
+#!/bin/bash
+# ComfyUI Startup Script
+
+set -e
+
+log() {
+    echo "[ComfyUI] $1"
+}
+
+# Download models using Python script
+python -u /workspace/download_models.py
+
+# Start ComfyUI
+log "Starting ComfyUI on port 8188..."
+exec python main.py --listen 0.0.0.0 --port 8188 "$@"