FLUX.1-dev-LoRA-AntiBlur

This is a functional LoRA trained on FLUX.1-dev for deep DoF (Anti-Blur🔥) by Vadim_Fedenko on Shakker AI. It may not be fancy, but it works.

Comparison

The following example shows a simple comparison with FLUX.1-dev under the same parameter setting.

It is worth noting that this LoRA has very little damage to image quality while enhancing the depth of field, and can be used together with other components, such as ControlNet. We regard it as a basic functional LoRA.

Trigger words

The trigger word is not required. The recommended scale is 1.0 to 1.5 in diffusers.

Inference

import torch
from diffusers import FluxPipeline

pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16)
pipe.load_lora_weights("Shakker-Labs/FLUX.1-dev-LoRA-AntiBlur", weight_name="FLUX-dev-lora-AntiBlur.safetensors")
pipe.fuse_lora(lora_scale=1.5)
pipe.to("cuda")

prompt = "a young college student, walking on the street, campus background, photography"

image = pipe(prompt, 
             num_inference_steps=24, 
             guidance_scale=3.5,
             width=768, height=1024,
            ).images[0]
image.save(f"example.png")

Online Inference

You can also run this model at Shakker AI, where we provide an online interface to generate images.

Acknowledgements

This model is trained by our copyrighted users Vadim_Fedenko. We release this model under permissions. The model follows flux-1-dev-non-commercial-license.