Open Creative Studio Nodes for ComfyUI

A custom node suite to augment the capabilities of Open Creative Studio for ComfyUI.

Nodes

Model Downloader v1

This node assists you in downloading an AI model from the Hugging Face repository containing all models used by Open Creative Studio.

You can specify the folder where you want to save a model, and the name the model will have in your folder.

If the folder structure you specify in the node doesn't exist, the node will create it for you.

If the file you specified already exists in the set folder, the node will ignore the request without errors.

The node supports absolute or relative folders, optional Bearer token authentication (you can reference an env var with $VARNAME), and the ComfyUI download progress bar if you are using the most recent versions of ComfyUI.

<img width="412" alt="Model Downloader v1" src="/Images/Model_Downloader_v1.png" />

Image Saver v1

This node allows you to save the input image/s in various formats: .png, .jpg/.jpeg, and .webp.

For each supported format, the node allows you to save a string of your preference in the EXIF tag UserComment (tag ID: 0x9286).

The UserComment tag can then be displayed by any image manipulation software supporting EXIF. Here's an example with XnView MP:

<img width="412" alt="EXIF UserComment in XnView MP" src="/Images/XnViewMP.png" />

If you choose the .png format, the node also allows you to embed the ComfyUI workflow in the extra section of the image metadata.

You can customize the filename with the following variables: %seed%, %date%, and %time%.

<img width="412" alt="Image Saver v1" src="/Images/Image_Saver_v1.png" />

Credit: This code is based on receyuki's SD Prompt Saver, available here, and willmiao's Save Image (LoraManager), available here. All credit to them.

Image Size (Local Models) v1

This node offers a list of preset resolutions for all local image generation models supported by OCS for ComfyUI: Black Forest Labs FLUX.1, Stability AI Stable Diffusion 3.5, XL, and 1.5.

The resolutions are organized by number of megapixels rather than by model as FLUX and SD3.5 are capable of generating both 1MP and 2MP images.

The user can also set a custom resolution.

<img width="412" alt="Image Size (Local Models) v1" src="/Images/Local_Image_Size_v1.png" /> <img width="412" alt="Image Size (Local Models) Menu v1" src="/Images/Local_Image_Size_Menu_v1.png" />

Image Size (Cloud Models) v1

This node offers a list of preset resolutions for all hosted image generation models supported by OCS for ComfyUI: OpenAI GPT-Image-1.

<img width="412" alt="Image Size (Cloud Models) v1" src="/Images/Cloud_Image_Size_v1.png" /> <img width="412" alt="Image Size (Cloud Models) Menu v1" src="/Images/Cloud_Image_Size_Menu_v1.png" />

Video Size (Local Models) v1

This node offers a list of preset resolutions for all local video generation models supported by OCS for ComfyUI: WanVideo 2.1, Hunyuan Video, CogVideoX 1.5 and 1.0.

The user can also set a custom resolution.

<img width="412" alt="Video Size (Local Models) v1" src="/Images/Local_Video_Size_v1.png" /> <img width="412" alt="Video Size (Local Models) Menu v1" src="/Images/Local_Video_Size_Menu_v1.png" />

Image List Filter v1

This node takes an image list as input, filters out images smaller than either width or height, and outputs a new image list without the excluded images.

If no input image remains after the filtering, the node outputs a new image list with the optional fallback image input.

<img width="412" alt="Image List Filter v1" src="/Images/Image_List_Filter_v1.png" />

Credit: This code is based on Kijai's Image Batch Filter, available here. All credit to him.

Installation

Install from ComfyUI Manager:

<img width="1378" alt="Open Creative Studio Nodes in ComfyUI Manager" src="/Images/ComfyUI_Manager.png" />

or clone this repo into the /comfyui/custom_nodes folder.