Deployment of the custom nodes, input files and workflows¶

Note: If you are running ComfyUI on GKE via ComfyUI on GKE guide, you can skip this guide and directly move to the User Guide.

ComfyUI custom nodes for Google Generative Media Models¶

ComfyUI custom node is an user-created extension that adds new functionality to the software's visual, node-based workflow.

Think of it like a plugin or a module. While ComfyUI comes with many built-in nodes, custom nodes allow users to implement new features, integrate external services, or combine existing functions into a single, more efficient node.

A custom node is typically a Python class that defines its inputs, outputs, and the function it performs. It can be as simple as a node that combines multiple images or as complex as a node that connects to an external API (like the Google Gemini API for image captioning) to process an input and return a new result.

In this guide we will use Google Generative Media custom codes named comfyui-google-genmedia-custom-nodes

Download Google Generative Media ComfyUI and other custom nodes¶

The demo contains workflows that uses Google Generative Media custom nodes. In addition to that, some of the workflows use custom nodes provided by ComfyUI-VideoHelperSuite for example to upload an input file from a given path. In this section, you will download the required custom nodes.

If you are running ComfyUI locally, perform the following steps to install the custom nodes¶

Download these custom nodes in one of the following ways:
- Using the ComfyUI Manager
  - Go to ComfyUI Manager --> Custom Node Manager --> search for comfyui-google-genmedia-custom-nodes. Click Install.
  - Go to ComfyUI Manager --> Custom Node Manager --> search for ComfyUI-VideoHelperSuite. Click Install.
- Manual installation using git
  - Open the terminal on the machine that is running ComfyUI and change to the directory where ComfyUI is running from.
```
cd <PATH_TO_COMFYUI_INSTALLATION>

E.g if your ComfyUI is running from "/home/user/username/ComfyUI", replace <PATH_TO_COMFYUI_INSTALLATION> with "/home/user/username/"
```
  - Change to the custom_nodes folder.
```
 cd ComfyUI/custom_nodes
```
  - Make sure git is installed on your machine.
```
git -v
```
    If you get the error command not found, install git based on your machine's operating system.
  - Clone the Google Generative Media custom node repository in the custom_nodes folder:
```
git clone https://github.com/GoogleCloudPlatform/comfyui-google-genmedia-custom-nodes
```
  - Clone the VHSHelper Suite custom node repository in the custom_nodes folder:
```
git clone https://github.com/kosinkadink/ComfyUI-VideoHelperSuite
```
  - Install python packages:
    1. If you are running ComfyUI in a python virtual environment, activate it before installing the dependencies.
      source <PATH_TO_PYTHON_VIRTUAL_ENVIRONMENT>/bin/activate
    2. Install python packages
      pip install -r comfyui-google-genmedia-custom-nodes/requirements.txt pip install -r ComfyUI-VideoHelperSuite/requirements.txt

Authentication to Google Cloud¶

Authenticate local ComfyUI instance with Google Cloud user¶

If you want to make the API calls to Google models with your Google Cloud user.

On the machine where you are running ComfyUI, install gcloud cli using the documentation https://docs.cloud.google.com/sdk/docs/install
Open the terminal.
Authenticate with the following command. Enter your Google Cloud credentials when prompted:
```
  gcloud auth application-default login
```

Authenticate local ComfyUI with Google Cloud Service Account¶

If you do not want to make the API calls to Google models with your Google Cloud user and want to use a service account instead.

On the machine where you are running ComfyUI, open the terminal.
Install gcloud cli using the documentation https://docs.cloud.google.com/sdk/docs/install
Authenticate with the following command. Enter your Google Cloud user credentials when prompted:
```
 gcloud auth login
```

Set shell variables

export PROJECT_ID="<YOUR_PROJECT_ID>"
export SERVICE_ACCOUNT_ID="<YOUR_SERVICE_ACCOUNT_ID>"
export USER_EMAIL=`gcloud auth list --filter=status:ACTIVE --format="value(account)"`

Replace <YOUR_PROJECT_ID> with the GCP project id you plan to use
Replace <YOUR_SERVICE_ACCOUNT_ID> with the name of the service account you will use to access Google Cloud with ComfyUI. It will be created in the next step.

Set Google Cloud project.
```
gcloud config set project $PROJECT_ID
```

Create a new service account:

gcloud iam service-accounts create $SERVICE_ACCOUNT_ID \
--display-name="ComfyUI Service Account" \
--description="Service Account to use ComfyUI"

Grant permissions to the service account:

Note: You need resourcemanager.projects.setIamPolicy permissions in order to run the following step. This permissions is included in owner , resourcemanager.projectIamAdmin and iam.securityAdmin roles.

gcloud projects add-iam-policy-binding $PROJECT_ID \
--member="serviceAccount:${SERVICE_ACCOUNT_ID}@${PROJECT_ID}.iam.gserviceaccount.com" \
--role="roles/storage.objectUser" --condition=None

gcloud projects add-iam-policy-binding $PROJECT_ID \
--member="serviceAccount:${SERVICE_ACCOUNT_ID}@${PROJECT_ID}.iam.gserviceaccount.com" \
--role="roles/aiplatform.user" --condition=None

Grant the Service Account Token Creator role to your user on the service account. This role lets your user account to impersonate the Service Account created in previous steps.

gcloud iam service-accounts add-iam-policy-binding \
${SERVICE_ACCOUNT_ID}@${PROJECT_ID}.iam.gserviceaccount.com \
--member="user:${USER_EMAIL}" \
--role="roles/iam.serviceAccountTokenCreator"

Now, set yourself to impersonate the service account.

gcloud auth application-default login --impersonate-service-account=${SERVICE_ACCOUNT_ID}@${PROJECT_ID}.iam.gserviceaccount.com

Copy the input images and workflows required for the demo¶

The input images and workflows are provided with this repo in the input and workflows directories respectively. You need to copy these in your ComfyUI instance in order to be able to run through this guide.

Open the terminal on the machine running ComfyUI.

Clone the git repository

git clone https://github.com/GoogleCloudPlatform/cloud-solutions /tmp/cloud-solutions

Set shell environment variables:
```
export YOUR_COMFYUI_INPUT_PATH="<YOUR_COMFYUI_INPUT_PATH>"
export YOUR_COMFYUI_WORKFLOW_PATH="<YOUR_COMFYUI_WORKFLOW_PATH>"
```
Note: Replace with the input folder of your ComfyUI instance e.g ComfyUI/input. Replace with the workflow folder of your ComfyUI instance e.g ComfyUI/user/default/workflows

Copy the input files

cp -r /tmp/cloud-solutions/projects/generative-media-shot-production/input/funday ${YOUR_COMFYUI_INPUT_PATH}

Verify that the input files have been copied over

ls -1d ${YOUR_COMFYUI_INPUT_PATH}/funday/*/*

You will see the following files. Note, that the path to the workflow directories could be different but you should see the same input files

/XXXXXX/ComfyUI/input/funday/casting/casting_reference_to_image.png
/XXXXXX/ComfyUI/input/funday/character_continuity/character_continuity_deflation.png
/XXXXXX/ComfyUI/input/funday/character_continuity/character_continuity_scene1.png
/XXXXXX/ComfyUI/input/funday/character_continuity/character_continuity_single_character_env.png
/XXXXXX/ComfyUI/input/funday/character_continuity/character_continuity_single_character_prop.png
/XXXXXX/ComfyUI/input/funday/environment_and_props/environment_and_props_reference_to_image.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production1_cast.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production1_environment.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production1_prop.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production2_cast.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production2_environment.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production2_prop.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production3_1.png
/XXXXXX/ComfyUI/input/funday/shot_production/shot_production3_2.png
/XXXXXX/ComfyUI/input/funday/storyboard/storyboard_cast1.png
/XXXXXX/ComfyUI/input/funday/storyboard/storyboard_cast2.png
/XXXXXX/ComfyUI/input/funday/storyboard/storyboard_environment1.png
/XXXXXX/ComfyUI/input/funday/storyboard/storyboard_environment2.png
/XXXXXX/ComfyUI/input/funday/storyboard/storyboard_final.png
/XXXXXX/ComfyUI/input/funday/storyboard/storyboard_prop1.png
/XXXXXX/ComfyUI/input/funday/storyboard/storyboard_prop2.png

Copy the workflow files

cp -r /tmp/cloud-solutions/projects/generative-media-shot-production/workflows/* ${YOUR_COMFYUI_WORKFLOW_PATH}

Verify that the workflow files have been copied over

ls -1 "${YOUR_COMFYUI_WORKFLOW_PATH}"/{casting,character_continuity,environment_and_props,storyboard,shot_production}/*

You will see the following files. Note, that the path to the workflow directories could be different but you should see the same json files.

/XXXXXX/ComfyUI/user/default/workflows/casting/casting_reference_to_image.json
/XXXXXX/ComfyUI/user/default/workflows/casting/casting_text_to_image.json
/XXXXXX/ComfyUI/user/default/workflows/character_continuity/character_continuity_deflation.json
/XXXXXX/ComfyUI/user/default/workflows/character_continuity/character_continuity_scene1.json
/XXXXXX/ComfyUI/user/default/workflows/environment_and_props/environment_and_props_reference_to_image.json
/XXXXXX/ComfyUI/user/default/workflows/environment_and_props/environment_and_props_text_to_image.json
/XXXXXX/ComfyUI/user/default/workflows/shot_production/shot_production1.json
/XXXXXX/ComfyUI/user/default/workflows/shot_production/shot_production2.json
/XXXXXX/ComfyUI/user/default/workflows/shot_production/shot_production3.json
/XXXXXX/ComfyUI/user/default/workflows/storyboard/reference_to_image_scene1.son
/XXXXXX/ComfyUI/user/default/workflows/storyboard/reference_to_image_scene2.json

Validate that the custom nodes are downloaded successfully¶

Re-start ComfyUI via ComfyUI manager or manually. Verify there are no failures in the logs and the log says that comfyui-google-genmedia-custom-nodes custom nodes have been loaded successfully, similar to the following message:
```
Import times for custom nodes:
 .....
   .....
   .....
   2.8 seconds: /XXX/XXXX/ComfyUI/custom_nodes/comfyui-google-genmedia-custom-nodes
```
Right click on ComfyUI, Click the 'Add node' menu and you should see a category named Google AI with custom nodes for Google’s genmedia models as shown in the image below.

Test a node to verify the connectivity with your GCP project¶

Open ComfyUI and right click on ComfyUI screen, follow the menu path by clicking Add node > Google AI > GeminiFLashImage > Gemini 3.1 Flash Image (🍌)
This will add a Gemini 3.1 FLash Image(Nano Bananan) node on the ComfyUI workflow screen. Enter your GCP project id in the gcp_project_id input field of the node and leave the region input as "global"
Write a prompt in the prompt input field of the node to generate an image.
Add your GCP project id and region to the node's respective input fields.
Right click on the ComfyUI screen and follow the menu path by clicking Add node > images > Preview Image
This will add a Preview Image node on the ComfyUI screen.
Now Connect the output of Gemini 3.1 FLash Image node to the input of Preview Image node. You should see something this the following imge:
Click the Run button on ComfyUI to run the workflow. It should generate an image and display it on the Preview Image node.

Note: You will need to pass a GCP project id and GCP region as input parameters to every custom node that is a part of google-comfyui-genmedia-custom-nodes suite.

Troubleshooting¶

If you see failures while testing a node, an error message will be displayed showing why the request failed.
If the error message displayed is not helpful, check the logs to see the errors.
If you are running the latest version of ComfyUI, click the Console menu on the bottom left of ComfyUI screen to see the logs.
Look for API errors. An example of a common error is
```
RuntimeError: Image generation API error: The project ID either doesn't exist or you don't have permissions to access it.
```
To fix it, verify that you are passing the correct GCP Project Id and you have permission to access it.