【 Workflow Introduction 】
To open the workflow, you need to upload two images on the left side, one base image and one style image, and use the "Runninghub Captioner" to reverse the base image prompt word, using the "LayerUtility: The "Get Main Colors" node extracts the main hue of the color card, then fine-controls the image and depth, and finally performs style transfer through Flux redux, which is combined with the sampler to generate the final image.
【 Usage Scenario 】
This workflow only needs two images, you can complete any style change, easy and efficient, built-in parameters are set, so use your imagination!

【 Key node 】
Flux redux

【 Model version 】
Flux
Model name :flux1-redux-dev.safetensors

【LoRA model 】
Model name :flux1-depth-dev-lora.safetensors

【Controlnet Application 】
There is no

【K Sampler 】
Sampling method: euler
Scheduler: simple
Noise reduction: 1
【 Custom Sampler 】
Sampling method: euler
Scheduler: simple
Noise reduction: 0.38