Enhanced run_comfyui.bat with Automated Dependency Checking and CUDA PyTorch Detection

[John-Caldwell]

Feature Request: Enhanced run_comfyui.bat with Automated Dependency Checking and CUDA PyTorch Detection

Problem

Windows users often encounter frustrating issues when setting up ComfyUI:

1. Missing Dependencies: Users encounter cryptic ModuleNotFoundError messages when dependencies are missing, requiring manual troubleshooting and installation
2. CPU-Only PyTorch: Many users accidentally install CPU-only PyTorch, which prevents GPU acceleration and causes significant performance issues without clear indication of the problem
3. Poor Error Messages: Existing error messages don't provide clear guidance on how to resolve issues, leaving users to search forums and documentation
4. Installation Confusion: Users are unsure which dependencies are required vs optional, and whether they should install in virtual environments

These issues create a poor first-time user experience and increase support burden.

Proposed Solution

Enhance the run_comfyui.bat startup script to:

• Automated Dependency Checking: Check all critical Python dependencies before launching ComfyUI, with clear prompts for missing packages
• CUDA PyTorch Detection: Automatically detect CPU-only PyTorch installations and offer to install the CUDA-enabled version
• User-Friendly Error Messages: Provide clear, actionable error messages with specific troubleshooting steps
• Virtual Environment Guidance: Detect virtual environments and provide appropriate warnings and guidance
• Progress Feedback: Show progress bars during installations for better user experience

Benefits

• Reduced Support Burden: Common setup issues are caught and resolved automatically
• Better User Experience: Windows users get clear guidance instead of cryptic errors
• GPU Support: Automatically ensures users have CUDA-enabled PyTorch for optimal performance
• Professional Appearance: Polished interface with clear formatting and helpful prompts

Implementation Details

The enhancement would:

• Check for missing dependencies using importlib.util.find_spec()
• Separate critical vs optional dependencies
• Detect CPU-only PyTorch by checking version string for "+cpu" indicator
• Provide interactive prompts for installation options
• Maintain full backward compatibility with existing functionality

Additional Notes

• All installations would be optional (users can cancel at any time)
• The script would warn users about system Python vs virtual environment implications
• All existing functionality would be preserved
• The enhancement is designed to be safe and non-destructive

Status

I have a complete PR ready to submit if this feature is desired. The implementation includes comprehensive dependency checking, CUDA PyTorch auto-installation, user-friendly error handling, and has been tested in various scenarios.

---

Note: This addresses common user pain points that may not have been formally reported as issues, but are frequently encountered in the community (especially on Discord/Matrix support channels).