Babes LoRA

Low-Rank Adapter for Flux. Dev trained on 100 to 170 babes images.

---

Why ?

I have no idea... but it works. Not only is it good at generating instagram-like images, it's also surprisingly great at realism - leading to better results than most flux realism based LoRA

Usage

Because of the little amount of data and the training techniques used, it is recommended to use these LoRAs with weights above 1.0. At these weights it does not mix very well with other LoRAs.

Weights are provided in safetensors format so they can be used with any generation library or platform.

Preview

Preview	Version	Recommended Usage/LoRA Weight
	-	Raw output from Flux Dev
	base	min=1.44, max=3.0
	1.1	min=1.44, max=3.0
	1.1a	min=2.0, max=2.44
	1.1b	min=2.0, max=2.44
	1.2a	min=1.44, max=2.0
	1.2b	min=1.44, max=3.0
	1.3a	min=1.44, max=2.0
	1.4a	min=1.44, max=2.0 Latest dataset

BibTeX

@Misc{babes,
  title =        {Babes: Flux Low-Rank Adapter trained on 100 babes images},
  author =       {Hugo Ventura},
  howpublished = {\url{https://huggingface.co/hugovntr/babes}},
  year =         {2025}
}