Hello there and thanks for checking out this workflow!
What's new in v27? : Major rework with new SubGraphs to streamline and compact the workflow even more, removal of outdated/broken nodes + fixes and optimizations all over!
—Purpose—
Built to provide an advanced, versatile and modular workflow for Flux with focus on efficiency, structure and information.
It comes with many notes explaining node settings and recommendations, as well as general guides from instructions to troubleshooting.
—Features—
Convenient loaders for all common versions of Flux and Clip models
Full metadata; recognized by CivitAI
LoRA support
SageAttention, EasyCache + Model Compile acceleration
ControlNet with Union support
Flux Tools LoRAs Canny + Depth (alt. to CNet based on LoRA)
Flux Redux (similar to IPAdapter)
PuLID (SVDQuant version only)
Wildcard prompting
Installation and download guide for models and nodes
multiple passes with optional upscales
— 1st : Detail Daemon + Variation Seed
— 2nd : DD. + Tiled Diffusion / UltimateSDUpscale
— ADetailer with dedicated LoRA Loader
— Inpainting
—Custom Nodes—
ComfyUI-nunchaku — SVDQuant version only
All of which can be installed through the ComfyUI-Manager
—Troubleshooting—
If nodes show up red (failing to load), check the 'Install Missing Custom Nodes' tab of the ComfyUI Manager for the missing node packs and install them.
Please check if all custom node packs load properly after installing, i.e. no
(IMPORT FAILED)messages next to any of them in the console upon ComfyUI startup.Always reload/drag'n'drop the original, downloaded workflow file into ComfyUI to reload an intact version of the workflow.
→ The last opened workflow that appears on startup shows a cached version of the workflow, "remembering" group nodes that failed due to missing nodes as failed, keeping them broken even after having everything installed correctly.
—Thanks—
The workflow would not be possible as is without these custom node packs. If you want to support the custom node creators, give them a ⭐ on their github repos! Thank you!
Feel free to ask questions, share improvements and suggestions in the comment section!
Let me know if you encounter confusing points I can elaborate on in the next update!
Description
v21
- upgrade of all guides and ►INFO boxes to markdown! 🎉
- design changes to make better use of highlight colors
- addition of dedicated LoRA Loader for ADetailer pass
- custom node pack changes :
- : Derfuu_ComfyUI_ModdedNodes
v21.1
- fixed connection error within "Sampler — Tiled Diffusion —OR— Ultimate Upscale" that resulted in Daemon Detail sampling only being applied to the U.Upscale option
FAQ
Comments (4)
Hi! I really like your workflow but it's got a few bugs to sort out. Here's one of them that turned me off from trying your workflow until I deep-dived the issue:
Bug: "Checkpoint Load — UNet - GGUF" group does not work correctly, error in console, creates job that cannot be canceled
Known bad versions: v21.1, v20
To reproduce:
1. Import v21.1 by dragging and dropping JSON into ComfyUI, grabbing latest version of all required custom nodes
2. Enable "Checkpoint Load — UNet - GGUF"
3. Disable "LoRA Settings", "2nd Pass"
3. Select models in GGUF section
4. Run workflow
Expected: Workflow runs to completion, image generated
Actual: Error in console (details below), workflow stuck, cannot cancel (inputs ignored)
Root cause: Unknown. Possibly due to dependencies being updated recently that broke one of the nodes within the GGUF groups?
Workaround:
1. Restart ComfyUI
2. Refresh page
3. Use the "Checkpoint Load — UNet" loader instead
Workaround 2:
1. Restart ComfyUI
2. Refresh page
3. Note down the fields, titles, etc. for each group
4. For all 4 group nodes in "Checkpoint Load — UNet - GGUF", right-click, choose "Convert to nodes"
5. Note down the values, titles, etc. for each node
6. Select every node from those four groups, right-click, choose "Fix node (recreate)"
7. Set the values, titles, etc. for each node as before
8. For each group, select, right-click, choose "Convert to Group Node"
9. Set the fields, titles, etc. for each group
10. Select models in GGUF section
11. Run workflow
Console output:
got prompt Failed to validate prompt for output 13: * ModelClamp 13957:0: - Exception when validating inner node: '14216:0' * ClipClamp 10596:0: - Exception when validating inner node: '14216:2' * VaeClamp 15567:1: - Exception when validating inner node: '14216:3' Output will be ignored Failed to validate prompt for output 11052:1: * GODMT_GetWidgetsValues 4488: - Required input is missing: ANY Output will be ignored Failed to validate prompt for output 11052:19: Output will be ignored Failed to validate prompt for output 4488: * (prompt): - Required input is missing: ANY Output will be ignored Failed to validate prompt for output 6308: Output will be ignored Exception in thread Thread-9 (prompt_worker): Traceback (most recent call last): File "C:\pinokio\bin\miniconda\lib\threading.py", line 1016, in _bootstrap_inner self.run() File "C:\pinokio\bin\miniconda\lib\threading.py", line 953, in run self._target(*self._args, **self._kwargs) File "C:\pinokio\api\comfy.git\app\main.py", line 175, in prompt_worker e.execute(item[2], prompt_id, item[3], item[4]) File "C:\pinokio\api\comfy.git\app\execution.py", line 496, in execute execution_list.add_node(node_id) File "C:\pinokio\api\comfy.git\app\comfy_execution\graph.py", line 122, in add_node inputs = self.dynprompt.get_node(unique_id)["inputs"] File "C:\pinokio\api\comfy.git\app\comfy_execution\graph.py", line 28, in get_node raise NodeNotFoundError(f"Node {node_id} not found") comfy_execution.graph.NodeNotFoundError: Node 14216:3 not foundHey there, thank you for pointing that out and reporting in detail! It's greatly appreciated! 🙏
I have not used any gguf models in a while, so I had to quickly download one to even reproduce the error. It seems like something has changed within the group node throughout the updates of Comfy and the node packs, but disassembling and rebuilding was enough to fix it, since nothing really changed.
Suspecting the same issue to affect the UNet NF4 loader as well, because of the same odd behaviour upon conversion to nodes. So I rebuilt that one as well.
Next update will be up tomorrow and include the fixes alongside several tweaks and upgrades 🙌
@RedPinkRetro Cool, thanks!
@athairus It will have to wait until tomorrow, as CivitAI is struggling to get its act together (as usual) 😒