
ComfyUI Flux Continuum - Modular Interface
A modular workflow that brings order to the chaos of image generation pipelines.
Updates
-
Flux Continuum Light 1.0.0:
-
Light version of the workflow with all the basic functions that requires only the FLUX.1-dev model. Download
-
-
1.4.5: Bug fixes
-
1.4.2: Black Forest Labs tools update
-
Black Forest Labs tools: Integrated the new Redux, Depth, Canny, Fill and IP Adapter models
-
Preview Panel: Preview all your image inputs and masks at a glance. Great for working with multiple IP Adapters
-
Mask Feather Control: Feather the mask using one control on inpainting, outpainting and detailer
-
Text Versions: Add more tabs via properties
-
New Nodes: FluxContinuumModelRouter, OutputGet, OutputGetString, OutputTextDisplay, DrawTextConfig and ConfigurableDrawText
-
-
1.3.0: Initial release
Overview
ComfyUI Flux Continuum revolutionizes workflow management through a thoughtful dual-interface design:
-
Front-end: A consistent control interface shared across all modules
-
Back-end: Powerful, modular architecture for customization and experimentation
✨ Core Features
Perfect for creators who want a consistent, streamlined experience across all image generation tasks, while maintaining the power to customize when needed.
-
Unified Control Interface
-
Single set of controls affects all relevant modules
-
No more juggling separate controls for each module
-
-
Smart Workflow Management
-
Only activates nodes and models required for current task
-
Toggle between different output types seamlessly
-
Efficiently handles resource allocation
-
-
Universal Model Integration
-
LoRAs and IP Adapterts work across all output modules
-
Seamless Black Forest Labs model support
-
-
Modular Architecture
-
Clear separation between front-end and back-end
-
Focus on creation in the front-end
-
Easily extend with new modules
-
? Quick Start
? New to Flux Continuum? Watch the tutorial first
-
Clone repo to the custom nodes folder
git clone https://github.com/robertvoy/ComfyUI-Flux-Continuum
-
Download Flux Continuum and import the workflow into ComfyUI
Flux Continuum Light is available here. (only FLUX.1-dev model is required)
-
Install missing custom nodes using the ComfyUI Manager
-
Configure your models in the config panel (press
2
to access) -
Download any missing models
-
Return to the main interface (press
1
) -
Select
txt2img
from the output selector (top left corner) -
Run the workflow to generate your first image
? Workflow Modules
All modules use the same unified control interface
Main Modules
txt2img Standard text-to-image generation. Generate images from prompts and tags
txt2img noise injection Enhanced detail generation for better results
img2img Load your image in the top-right corner and adjust the Denoise slider
inpainting Mask-based image editing with Black Forest Labs Fill model integration
outpainting Expand images beyond their boundaries with controllable padding. Uses the Black Forest Labs Fill model.
canny Control generation with the Black Forest Labs canny model using edge detection
depth Control generation with the Black Forest Labs depth model
detailer Focused refinement using mask selection. Use the Img Mask tab to create/load mask
ultimate upscaler Advanced upscaling with model selection, denoising, and resolution control
upscaler Simple model-based upscaling with resolution control
face swap Replace faces between images. Access settings by pressing 3 on keyboard
Utility Modules
imgload prep Preview processed images after crop/sharpen/resize/padding
canny preprocessor Preview and adjust canny edge detection settings
depth preprocessor Configure and preview depth map generation
? Custom Nodes
These custom nodes were made specifically for this workflow and are required for it to work
Interface Enhancement Nodes
-
Tabs:
-
Space-saving node organization
-
Add tabs via the properties panel
-
Compatible with most nodes
-
Special handling for unsupported nodes (e.g., Rgthree) using dual-tab workaround
-
-
Sliders:
-
Suite of pre-configured sliders
-
Optimized ranges and defaults for common operations
-
Seamless integration with Flux workflow
-
-
OutputGet
-
Filters set nodes with the prefix
Output -
-
Comes with OutputTextDisplay: displays the name of the selected output in the OutputGet node
-
And OutputGetString: outputs the selected output name in the OutputGet node as a string
-
-
Text Versions:
-
5-tab text management system for prompt organization
-
Add more tabs using the properties panel for this node
-
Save different versions of your prompt/text
-
Great for experimenting or using the detailer/inpainting
-
-
ImageDisplay Extension:
-
Displays Base64 images directly on canvas
-
Configurable via properties panel
-
Use sparingly to maintain optimal file size
-
Parameter Management Nodes
-
Sampler Parameter Packer: Consolidates sampler settings with tabbed interface
-
Sampler Parameter Unpacker: Extracts packed parameters for workflow integration
-
Batch Boolean: Smart batch processing control. Adds the second input image to the batch only if conditon is met
-
Flux Continuum Model Router: Intelligent model selection and routing. Designed specifically for this workflow
-
Configurable Draw Text: Takes an input from a Draw Text Config node with text style settings and renders text on top of an image
-
Value Pass: An extension of
ComfyUI-KJNodes
pass-through functionality for Latent, Pipe, and SEGS data
? Coming Soon
-
Multi-GPU Support: Processing with multiple GPUs
? Acknowledgments
Special thanks to the creators of these essential custom node packs:
描述:
What's New in 1.4.2:
- Black Forest Labs Redux, Depth, Canny, Fill models and IP Adapter integration
- New preview panel to monitor all your inputs at once (great for working with multiple IP Adapters!)
- Enhanced mask controls for inpainting, outpainting, and detailer
- More customization options for Text Versions node
- New utility nodes for better workflow control
训练词语:
名称: fluxContinuum_v144.zip
大小 (KB): 498
类型: Archive
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success