LogoWonderful Launcher
  • Product
  • Pricing
  • Solutions
  • Docs
  • Blog
  • Get Help
  • Download

How to Fix Red Nodes in ComfyUI Workflows

Needs verification

Diagnose and fix red nodes in ComfyUI workflows caused by missing custom nodes, import failures, or version incompatibilities.

Community Knowledge

This page is based on common ComfyUI troubleshooting patterns and has not been fully tested across all environments. Back up your environment before changing packages.

Red nodes in a ComfyUI workflow mean the node type exists in the workflow file but is not registered in your current ComfyUI session. The node cannot execute, and any workflow containing red nodes will fail when queued.

This guide covers all the reasons nodes turn red and how to fix each one.

Fast answer

Red nodes have three main causes:

  1. Custom node not installed — install the missing node pack
  2. Custom node installed but import failed — fix the dependency error
  3. Node type renamed or removed — update the node pack or replace the node

What red nodes mean

A red node means ComfyUI cannot find the Python class that defines that node type. The workflow JSON references a node by its type name (e.g., "IPAdapterApply"), but no installed node registered that name during startup.

Red nodes are different from:

  • Yellow nodes: Usually indicate a warning or optional connection issue
  • Nodes with error badges: The node is loaded but encountered a runtime error
  • Greyed-out nodes: The node is muted/bypassed

Why nodes turn red

Cause 1: Custom node pack not installed

The most common case. The workflow was created on a system that had custom nodes you do not have.

Fix: Install the missing custom node pack. See Workflow Missing Nodes for detailed steps.

Cause 2: Custom node installed but failed to import

The node pack folder exists in custom_nodes/, but Python could not load it. Check the ComfyUI startup log for lines like:

IMPORT FAILED: ComfyUI-SomeNodePack
ModuleNotFoundError: No module named 'somepackage'

Fix: Install the missing dependency:

.\python_embeded\python.exe -s -m pip install somepackage

Then restart ComfyUI.

Cause 3: Node type was renamed or removed

Custom node authors sometimes rename node classes in updates. If the workflow references the old name, it will show red.

Fix:

  • Update the custom node pack: cd ComfyUI\custom_nodes\<pack> && git pull
  • If the node was renamed, you may need to re-add the node manually in the workflow and reconnect its inputs/outputs
  • Check the node pack's changelog or GitHub issues for migration instructions

Cause 4: ComfyUI version mismatch

Some built-in nodes were added in newer ComfyUI versions. If your ComfyUI is outdated, newer workflows may reference nodes you do not have.

Fix: Update ComfyUI:

cd ComfyUI
git pull

Do not run pip install -r requirements.txt blindly.

  1. Confirm which Python environment is active (where python or which python)
  2. Inspect requirements.txt — check if it pins torch, numpy, or opencv versions
  3. Install only the missing package when possible: pip install <package-name>
  4. Run pip check after installation to verify no conflicts were introduced

Cause 5: Conflicting node packs

Two custom node packs may register conflicting node names, or one may interfere with another's imports.

Fix: Check the startup log for IMPORT FAILED and resolve conflicts. Try disabling one pack at a time to isolate the issue.

How to diagnose red nodes systematically

Step 1: Note all red node names

Click on each red node and note the type name shown. Or load the workflow JSON and search for all unique type values.

Step 2: Check the startup log

Look for IMPORT FAILED messages. Each one tells you which custom node pack failed and why.

Step 3: Categorize the failures

Red node causeWhat the log showsFix
Not installedNo mention of the packInstall the pack
Import failed - missing packageModuleNotFoundErrorInstall the missing pip package
Import failed - version conflictImportError or AttributeErrorFix the dependency or update the pack
Renamed nodePack loads but node name does not matchUpdate pack or edit workflow JSON

Step 4: Fix in priority order

Fix import failures first (they may resolve multiple red nodes at once), then install missing packs, then handle renamed nodes last.

Editing workflow JSON as a last resort

If a node was renamed and you know the new name, you can edit the workflow JSON directly:

  1. Export or save the workflow as a JSON file
  2. Open it in a text editor
  3. Find and replace the old node type name with the new one
  4. Reload the workflow in ComfyUI

Only do this if you are confident about the correct new node name.

When not to run pip install blindly

  • Red nodes do not always mean you need to install more packages — the node pack itself may need to be installed first
  • Do not install every suggested package from every node pack at once
  • If a node pack is very old and unmaintained, consider finding a modern alternative rather than fighting dependency issues

How Wonderful Launcher can help

Wonderful Launcher helps identify red nodes and their root causes. It can help distinguish between missing installations and import failures, and guides you through the correct fix for each case.

Download Wonderful Launcher — it's free and simplifies workflow troubleshooting.

Related errors

  • Workflow Missing Nodes
  • Nodes Show Red After Importing Workflow
  • ComfyUI Plugin Import Failed
  • ComfyUI Dependency Conflicts
  • ComfyUI Startup Failed

Source References

  • ComfyUI GitHub
  • ComfyUI Documentation
  • ComfyUI Manager GitHub

You can fix it manually, or download Wonderful Launcher for Windows to diagnose plugin errors, missing dependencies, and broken ComfyUI environments without reinstalling.

Download free for Windows

Did this fix your issue?

Your answer helps prioritize verified ComfyUI repairs.

Table of Contents

What red nodes mean
Why nodes turn red
Cause 1: Custom node pack not installed
Cause 2: Custom node installed but failed to import
Cause 3: Node type was renamed or removed
Cause 4: ComfyUI version mismatch
Cause 5: Conflicting node packs
How to diagnose red nodes systematically
Step 1: Note all red node names
Step 2: Check the startup log
Step 3: Categorize the failures
Step 4: Fix in priority order
Editing workflow JSON as a last resort
When not to run pip install blindly
How Wonderful Launcher can help
Related errors
Source References