Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
This is designed for Klein (or flux2), but the models & scheduler can be changed for other edit models.
Runtime profile
Source description
This is designed for Klein (or flux2), but the models & scheduler can be changed for other edit models.
This was designed with the non-base (turbo/distilled) version of Klein in mind, though you can also use base+turbo. non-base Klein is 4 step turbo model, or can use the turbo lora - use with 1 CFG. If using base make sure to increase steps to 20+ & CFG to 4+
This is not exactly inpainting, but with Klein, it sometimes is better than inpainting and it is much faster.
Masked or Unmasked - masking is way faster - crops down to rectangle then composites back into your mask. Great for making small tweaks. recommend 1 hardness or it might resize in the transparent areas. Masked editing only takes like 2.5s on my 5080. unmasked depends on image size and better when you want to change the entire image
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
This workflow is for image-to-image editing—not exactly inpainting: provide an image, editing prompt, and optional mask; edit a masked crop and composite it back or edit the whole image, producing original and edited versions in the same size and file type with optional color
With a mask, it crops the edit to a rectangle and composites the result back into the masked area; without a mask, it can edit the whole image.
Use masked editing for small tweaks and unmasked editing when you want to change the entire image.
Suggestion · not verified
If the cropped area is too small, try upscaling before sampling; if an unmasked image is too large, try downscaling.
Suggestion · not verified
For I2I datasets, it outputs the original and edited versions in the same size and file type; optional color matching can help keep color consistent between them.
It is published by Epispasm as v1.0 and is designed for Flux.2 Klein 9B-base; its models and scheduler can be changed for other edit models.
License information is not specified for this workflow.
The workflow lists these model files: flux-2-klein-9b-fp8.safetensors; Flux2 Klein动漫转写实真人 AnythingtoRealCharacters.safetensors; flux2-vae.safetensors; klein_9B_Turbo_r128.safetensors; qwen_3_8b_fp8mixed.safetensors.
The listed node packs are: comfyui_essentials (1.1.0); comfyui-crystools (1.27.4); comfyui-impact-pack (8.28.2); comfyui-inpainteasy (1.0.2); comfyui-kjnodes (1.2.9); comfyui-sam2 (1.0.3); RES4LYF (0dc91c00c4c3fb38e7874fcd7a2a327765e8882c); rgthree-comfy (1.0.2512112053); and was
For non-base turbo or distilled Klein, use 1 CFG; with the base model, use 20+ steps and 4+ CFG.
Suggestion · not verified
For masked editing, set mask hardness to 1; otherwise transparent areas might be resized.
Suggestion · not verified
Turn color match on when you want to bring colors back toward the original, and off for style changes or black-and-white edits.
Suggestion · not verified
Before using it, verify the exact flux2-vae.safetensors file and make sure was-ns is available.
Suggestion · not verified
Image-edit models may shift subjects in the composition; to seek a closer match, try a finetune or LoRA, adjust the prompt, expand the mask, or increase batch size.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
3 excerptsSource context: 561 downloads · Type Workflows · Base model Flux.2 Klein 9B-base
Upscaled or not before sample - may help with additional details if your cropped area is too small. or if using unmaksed you can downscale if your image is too big. I usually leave it off.
color match or no color match - definitely a godsend if you are trying to inpaint and you want to bring the color back to the original. Helps gets rid of inpaint auras. turn off if you don't need to match the original (changing styles, to black & white, etc
PURPOSE
Fast I2I image editing - edit just a small crop and composite it back or edit the whole image, all with one workflow.
This workflow is great for creating new I2I datasets - it will output the original and edited version in the same size and filetype, and optional color match helps ensure that color is consistent between both versions.
Note that image edit models by themselves do not always keep subjects in the same composition. finetunes and loras can help prevent subjects from shifting. You can tweak your prompt, expand your mask, or increase batch size to get a closer match.
Remaining hardware or download-size estimates are not available yet.
Requirements
Checkpoint · 8.79 GB · SAFETENSORS · Hugging Face · black-forest-labs/FLUX.2-klein-9b-fp8
Flux2 Klein动漫转写实真人 AnythingtoRealCharacters.safetensors
File unverifiedSource context: 549 downloads · Type Workflows · Base model Flux.2 Klein 9B-base
Source context: 535 downloads · Type Workflows · Base model Flux.2 Klein 9B-base
LoRA · Hugging Face · WarmBloodAban/Flux2_Klein_Anything_to_Real_Characters
VAE · 321 MB · SAFETENSORS · fp8 · pruned · Hugging Face · Comfy-Org/flux2-dev
LoRA · 1.29 GB · SAFETENSORS · Civitai · 2324315
Text encoder · 8.07 GB · SAFETENSORS · fp16 · Hugging Face · Comfy-Org/flux2-klein-9B
Node pack · Registry
comfyui-crystools
FoundNode pack · Registry
comfyui-impact-pack
FoundNode pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry
RES4LYF
FoundNode pack · Registry
rgthree-comfy
FoundNode pack · Registry
was-ns
Not resolvedNode pack · Registry
