Dutch Blonde influencer – Woman037– Character LoRA (Perfect for Fanvue) [SDXL | SD 1.5 | Flux]

Use the LoRA Loader (Z-image) (CRT) node.

For more consistency you can add: blonde hair

Note : This is my first experimental test with Z-image Turbo training, made with help from AI Toolkit. More updates will follow as I improve the workflow. As seen in all my gallery examples, the image and character can still be heavily influenced by the prompt. A short prompt keeps it more faithful to the LoRA, but it's fun to experiment with long prompts via an LLM.

Settings for the example images Workflow: https://comfyanonymous.github.io/ComfyUI_examples/z_image/ ModelSamplingAuraFlow: 7 Steps: 16 CFG: 1.0 Sampler: euler Scheduler: sgm_uniform denoise: 1

All the examples posted are raw output.