Person Enhancer - FLUX Style (for SDXL/PONY)

This LoRA adds FLUX-like details to images of people in SDXL and PONY. It was trained on SDXL base, but works in PONY realism models too. FLUX shows some distinct characteristics in skin realism, hands, lips, and body structure. This LoRA lets you achieve some of that in SDXL/PONY checkpoints.

The model was trained on 764 dual pass images from flux dev. It is a dataset I am building to train into checkpoints. All photos contain images of people with randomized gender, clothing, activity, ethnicity, etc. So the LoRA is definitely targeted toward realistic images of people. I wanted to create a LoRA as a proof of concept and to measure results/future needs. Once I tested and found the results were good (great even), I decided to release the LoRA now. Once I build out the remaining data set, I may update it again.

I tested this on a few different models with results in the post. You can see the different models in the image data. There was no adetailer applied - just hiresfix and remacri upscale.

Note: I have not tested its impact on images without people in them. I would guess it would have some by nature of the detailed backgrounds in FLUX, but it was not the areas I was targeting in this training run.

For best results, use the triggerword "fluxenh4nce". It works without it, but less impactful. Strength of 0.5-1.0 depending on prompt length and checkpoint. I used 0.8 the most.

Here is a sample image without the LoRA applied:

And this is with it applied (same prompt, seed, checkpoint, etc:Disclaimer: This model is provided for educational purposes and for the advancement of AI art in general. The creator of this checkpoint assumes no responsibility or liability for the end user's outputs. Any illegal, unethical, or deep fake use of this model is the sole responsibility of the user engaged in those activities. The end user assumes all liabilities from use of the checkpoint.

No permissions are given for anyone to distribute this model on other sites. I do not give permission for it to be used on any image generation service other than CIVITAI.

描述:

Based on 764 highly detailed FLUX images of people.

训练词语: fluxenh4nce