Flux DreamBooth LoRA - weathon/anti_aesthetics_lora

Project Page: https://weathon.github.io./Anti-aesthetics-website/ Paper: https://arxiv.org/abs/2512.11883

Model description

These are weathon/anti_aesthetics_lora DreamBooth LoRA weights for black-forest-labs/FLUX.1-dev.

The weights were trained using DreamBooth with the Flux diffusers trainer.

It is trained on this dataset https://huggingface.co/datasets/weathon/aas_distillation

Trigger words

You should use any words that describe anti-aesthetics to trigger the generation. Examples are shown in the training set https://huggingface.co/datasets/weathon/aas_distillation in the re_prompted column.

Download model

Download the *.safetensors LoRA in the Files & versions tab.

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers

Usage

You will HAVE TO use this LoRA with Flux Guidance scale (not CFG scale) of 3.5. And it is also recommonded to set lora_scale above 1 if you want max anti-aesthetics effects.