nodes/LaoMaoBoss-ComfyUI-WBLESS
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
108 Commits 1 Branch 28 Tags
main
T
Clone
Open with VS Code Open with VSCodium Open with Intellij IDEA
Download ZIP Download TAR.GZ Download BUNDLE

README.md

ComfyUI-WBLESS

ComfyUI custom node package. This custom node features multiple practical functions, including global variables, flow control, obtaining image or mask dimensions, and Dominant Axis Scale.

Get Started

Recommended

Manual

  • Clone this repo into : custom_nodes 
    cd ComfyUI/custom_nodes
git clone https://github.com/LaoMaoBoss/ComfyUI-WBLESS.git
  • Start up ComfyUI.

NOTICE

  • V3.5.0 Add the ToAPIs GPT Image 2 VIP node.
  • V3.4.3 Fix Python 3.13/Lexicon loading, loguru fallback, and Jimeng SDK import errors.
  • V3.4.2 RunningHUB LLM node supports multiple image inputs.
  • V3.4.1 RunningHub LLM node adds image input.
  • V3.4.0 Add RunningHUB LLM Node.
  • V3.3.0 Add the RunningHub DeepSeek Chat node and update the DeepSeek Chat node.
  • V3.2.0 Update Qwen API Node.
  • V3.1.0 Update DeepSeek API Node.
  • V3.0.1 Update image understanding capabilities.
  • V3.0.0 Add Gemini Chat node.
  • V2.9.0 Update RunningHUB API, API Core-RustFS Node.
  • V2.8.0 Add API Core Node.
  • V2.7.0 Add Image Mask Blend Node.
  • V2.6.0 Add the Jimeng Image 4.0 node and enhance the functionality of the Text Block node.
  • V2.5.0 Added Baseline Alignment (X) and Baseline Alignment (Y) nodes.
  • V2.4.2 Fix the calculation error issue of the Area Based Scale (Pixel) node.
  • V2.4.1 Restore the name Area Based Scale (Size) to Area Based Scale.
  • V2.4.0 Add ImageHasAlpha and Area Based Scale (Pixel) nodes,Rename the Area Based Scale node to Area Based Scale (Size).
  • V2.3.1 Add an ApplyMaskToAlpha node and add new settings to the Gradient node.
  • V2.3.0 Add Gradient Node.
  • V2.2.4 Fix the adaptation of the switch series nodes in the new version of ComfyUI.
  • V2.2.3 Fix switch node runtime errors in certain environments.
  • V2.2.2 Update the Overlay Text and Text Block nodes.
  • V2.2.0 Update the Area Based Scale Node.
  • V2.1.2 Import the cozy_comfyui module and fix the issue where users cannot import the cozy_comfyui module.
  • V2.1.1 Fixed the Get Mask Size node returning wrong dimensions in certain scenarios.
  • V2.1.0 Updated the Dominant Axis Scale, Get Image Size, and Get Mask Size nodes.
  • V2.0.0 is released. This is the first public version. Version 1.0 is an internal test version and will not be made public.

The Nodes

Set Global Variable

The Set Global Variable node allows you to store your data in variables.

See More Information
  • The Input and Output nodes form a direct pipeline for better integration within workflows.
  • The variable data is used for inputting variable values.
  • Scope is used to set the order in which variables are obtained. You just need to connect them in sequence one after another.
  • variable_name Here you can set the name of your variable. image

Get Global Variable

The Get Global Variable node can retrieve data stored in variables.

See More Information
  • The Input and Output nodes form a direct pipeline for better integration within workflows.
  • variable data is used for outputting the variable's value.
  • Scope is used to set the order in which variables are obtained. You just need to connect them in sequence one after another.
  • variable_name Here you can specify the variable you want to retrieve. image

Inversed Switch

Used to control the direction of the workflow.

See More Information
  • Connect the main workflow to the Input interface, then connect the Output to different branch workflows. By controlling the path value of the node, you can determine which branch the workflow will take.
  • This node needs to be used in conjunction with Switch.
  • The core logic of this node draws inspiration from ComfyUI-Impact-Pack. We would like to express our gratitude to the author of ComfyUI-Impact-Pack here. workflow

Switch

Select and retrieve data from different processes.

See More Information
  • This node is usually used in conjunction with the Inversed Switch node; of course, you can also use it independently.
  • The Input interface connects to different branch workflows, while the Output interface will output data from the corresponding workflow based on the value of path. image

Dominant Axis Scale

Smart scale input group A relative to input group B, using its longest side as the scaling axis.

See More Information
  • Height a, Width a — these are the input dimensions you need to scale.
  • Height b, Width b — these reference dimensions serve as the scaling baseline, which you can conceptualize as canvas dimensions.
  • ratio — Input your scaling factor here.
  • The output Width, Height, and scale_ratio govern different output formats. workflow (1)

Area Based Scale

Smart scale the area of input group A with reference to input group B, based on size.

See More Information
  • Height a, Width a — these are the input dimensions you need to scale.
  • Height b, Width b — these reference dimensions serve as the scaling baseline, which you can conceptualize as canvas dimensions.
  • ratio — Input your scaling factor here.
  • The output Width, Height, and scale_ratio govern different output formats.
  • cap_threshold — the upper scaling limit threshold, beyond which the object will not scale any.
  • enable_cap — threshold activation switch. workflow (4)

Area Based Scale (Pixel)

Smart scale the area of input group A with reference to input group B, based on Pixel.

See More Information
  • The image_alpha port connects to an image with a transparency channel. You may need to use it in conjunction with ApplyMaskToAlpha. Unless you explicitly clear the transparency information of the image, please use it alongside ApplyMaskToAlpha to generate an image with transparency channel information. For specific usage, refer to the ApplyMaskToAlpha section.
  • The image input port is used to connect the background image, or it can be referred to as the reference for scaling. Area Based Scale (Pixel)

Get Image Size

Get Image Dimensions.

See More Information image

Get Mask Size

Get Mask Dimensions.

See More Information image

Overlay Text & Text Block

The function of drawing text on images consists of two nodes: a parent node and a child node. The parent node is Overlay Text, which is used to control the overall settings, while the child node is Text Block, responsible for controlling individual text blocks. These two nodes work together to build a powerful text system with professional-level text adjustment functions.

See More Information
  • The text_block input of Overlay Text is used to connect text blocks. If multiple styles need to be controlled separately, connect multiple text blocks. workflow (4)

Gradient

Used to generate gradients or transparent gradients for images.

See More Information
  • gradient_type is used to set the gradient style.
  • rotation_angle is used to set the gradient direction.
  • The position series settings are used to control the gradient effect in detail.
  • The color series settings are used to set gradient colors.
  • The alpha series settings are used to set gradient transparency.
  • The mask output port will output the corresponding mask based on the transparent gradient. workflow

ApplyMaskToAlpha

Set the image transparency based on the mask.

See More Information
  • Connect the mask information to the mask input, and it will output an image with an alpha channel. workflow

ImageHasAlpha

Determine whether the image contains transparency channel information.

See More Information ImageHasAlpha

Baseline Alignment (X) & Baseline Alignment (Y)

Input two different heights, then set a baseline. The output will be how much the center position of height a needs to be adjusted so that its bottom position aligns with the baseline position of height b.

See More Information Baseline_Alignment_demo

Jimeng Image 4.0

Jimeng Image 4.0 node.

See More Information
  • Please go to 火山引擎 to complete the key application.
  • Since Jimeng 4.0 does not currently support base64, please go to PicGo to obtain the image hosting key. workflow (5)

Image Mask Blend

Input a background image and an image to be blended, then input the mask of the area in the background image where blending is required. The node will scale and move the layer image based on the size and position of the mask.

See More Information workflow (1)

API Core

API Core Transfer Node.

See More Information
  • Please go to API Core to complete the key application.
  • Please fill in your ComfyUI web address in server_origin for uploading reference images. Note that this address must be a public IP address or domain name.
  • Once you fill in the api_key, the node will automatically retrieve available models for you to select. image

API Core-RustFS

API Core variant that uses RustFS to upload image URLs, and requires self-deployment of RustFS.

See More Information image

RunningHUB API

RunningHUB API Invocation Node.

See More Information image

Gemini Chat

Gemini Chat Invocation Node.

See More Information image

DeepSeek Chat

DeepSeek API Node.

See More Information image

Qwen Chat

Qwen API Node.

See More Information image

RunningHub DeepSeek Chat

RunningHub DeepSeek Relay Node.

See More Information
  • Please visit RunningHub to apply for an API key.
  • If you are a user from China, you can set the node’s base_url parameter to https://llm.runninghub.cn/v1 and go to RunningHub to apply for an API key. image

RunningHub LLM

RunningHub LLM Relay Node.

See More Information
  • Please visit RunningHub to apply for an API key.
  • If you are a user from China, you can set the node’s base_url parameter to https://llm.runninghub.cn/v1 and go to RunningHub to apply for an API key. image

ToAPIs GPT Image 2 VIP

ToAPIs GPT Image 2 VIP image generation node. Supports text-to-image, single-reference and multi-reference image-to-image.

See More Information
  • Please go to ToAPIs to apply for an API key. The api_key field can be left empty if the TOAPIS_API_KEY environment variable is set.
  • api_base defaults to https://toapis.com. Users in mainland China should change it to https://toapis.cn.
  • The size list covers every preset ratio (1:1, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 16:9, 9:16, 2:1, 1:2, 21:9, 9:21, plus auto). For an arbitrary ratio such as 7:4 or 1:3, fill in custom_size; it overrides the preset and must stay within a 3:1 aspect ratio.
  • resolution (1k / 2k / 4k) decides the actual pixel size, while quality (low / medium / high) trades speed for fidelity.
  • Reference images must be public http(s) URLs, so local ComfyUI images are uploaded to the ToAPIs /v1/uploads/images endpoint automatically. Each file is limited to 10MB, and an oversized PNG falls back to JPEG.
  • The generation interface is asynchronous. The node submits a task and then polls it, so timeout and poll_interval control the total wait and the polling cadence. image
Reference in New Issue View Git Blame Copy Permalink
S
Description
No description provided
Readme MIT
960 KiB
Languages
Python 92.5%
JavaScript 7.5%