My TG Channel - https://t.me/StefanFalkokAI
My TG Chat - https://t.me/+y4R5JybDZcFjMjFi
Hi! I introduce my working workflows with Flux.2 Klein 9B for ComfyUI.
I have included workflows with Flux.2 Klein 9B models
You need to have Flux.2 Klein 9B KV Model (https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-kv/blob/main/flux-2-klein-9b-kv.safetensors), clip (https://huggingface.co/Comfy-Org/vae-text-encorder-for-flux-klein-9b/tree/main/split_files/text_encoders) and vae (https://huggingface.co/Comfy-Org/flux2-dev/tree/main/split_files/vae)
if you need Base version - https://huggingface.co/black-forest-labs/FLUX.2-klein-base-9B/blob/main/flux-2-klein-base-9b.safetensors
also recommend download Flux 2 Klein 9B by wikeeyang, it gives a better result - https://civarchive.com/models/2339723/flux2-klein-9b-true
https://huggingface.co/datasets/StefanFalkok/ComfyUI_portable_torch_2.13.0_cu130_cp313_sageattention_triton - my new ComfyUI build with torch 2.13.0cu130 + sageattention + my workflows, also u need download and install CUDA 13.0 (https://developer.nvidia.com/cuda-13-0-0-download-archive) and VS Code (https://visualstudio.microsoft.com/downloads/)
Leave comments if you have trouble or you found the problem with workflows
Description
minor update - upgraded inpaint workflow
FAQ
Comments (2)
The workflow works and the results are really impressive. I especially like the True model it produces more natural and realistic faces. The KV model also has its own strengths, but the results feel more like a polished/glamour retouched portrait.
However, the biggest downside for me is the lack of control, especially in the Edit workflow. When using image editing, the guidance seems to mostly affect prompt adherence, but there is no clear way to adjust how strongly the original image should be preserved.
This is a very important control for my use case, because the workflow can completely change the original image: the camera angle, head direction, body position, composition, etc. Sometimes the result is beautiful, but it becomes a completely different image.
I’m not looking for a full reinterpretation of the picture. I need the model to keep the original pose, lighting, and composition, while improving the face, skin texture, and realism.
The model itself is amazing, but without a proper image strength/denoise control, it is difficult to use for precise editing where preserving the original image is important.
Then you can try using krea 2 edit I've updated










