Flux.2 Klein Mix Workflow
Workflow with main focus on Image Editing, with control over Image Sizes. Trying to only use the more common Custom Nodes or ones that do not have requirements for a easy and safe installation.
Aiming for consistency and support edits for up to 4 images and Text to Image, 3 different Controlnets, different Inpainting methods and Outpainting. Nodes for selecting image sizes and Aspect Ratio. Uses Flux2 Klein Enhancer nodes, for consistency and prompt impact. Color matching and different sampler with ETA setting.
The workflow has many notes that explain things inside, which you can read or delete.
There is a second Version inside for people that use Lora Manager so you can pick what you want.
The workflow only support a Positive Prompt except for the 2-pass mode in V5 for T2I that uses both the Base and the Normal model and a negative prompt.
All custom nodes used:
Essentials
Rgthree
Easy-Use
Impact Pack
KjNodes
GGUF
Controlnet_aux
LayerStyle
ArtVenture
Below Does not have any requirements making for an easier and safer install:
Inpaint-CropAndStitch
ModdedNodes
Comfyroll
ControlAltAI
Flux2Klein-Enhancer
CRZnodes
mxToolkit
ZImageTurbo-FlowSampler
Note that V2 and V1 do not use the same custom nodes, and are harder to use because of that, and I do not recommend using them!
Feel free to criticize, comment or ask questions.
Description
Almost a complete remake.
Uses more commonly used or easily installed custom nodes.
One more Enhancer added and moved for easier overview.
Many settings and nodes changed inside sub-graphs.
Image size selection changed, and all now uses value 0.0 to keep the original.
Generation Settings node added.
Control Net moved to Image 2.
Color Matching added.
Negative Prompt from v2 removed.
Small cropping added to everything except in-painting if the sizes is not already divisible with 16 that the model require.
GGUF-Safetensor choice do no longer require both to have a model selected.
FAQ
Comments (6)
The I2I is constantly making noise. Attempts to find the problem have been unsuccessful. Has anyone else experienced the same thing? What should I do?
Unsure what kind of noise you are talking about but in the "Ref Latent Controller" there is a setting for "blend_with_noise" It should have been set to 0.05 when you downloaded the workflow which should not be noticeable and I found the picture sometimes a little better then at 0.00. Maybe you have increased that?
@Whistler42 After and during generation, it shows noise. There's not even a hint of an image. The sampler is the same.
@Ashaf I have not encounter that so do not know what is wrong. Do you get any errors in the log? Does T2I work? If it does not work maybe you have loaded the wrong model you need 3 of them correct, or you have some lora that does not work loaded... You can try and download the workflow again also in case you messed something up.
Hey man really trying to get this work and all I can see is grey/noise in outputs. do you know whats going on with it. I seem to be using the same model/vae and everything you suggest.
Second one to say that and I do not know why. I tried to recreate it but have not been able to. Hard if you do not get any error message also and it works for me... I am working on V4 that is soon done. Maybe I accidentally fixed the that error then. It does have links to the correct models even if you said you used the correct ones.