---
title: FLUX.1-dev-LoRA-AntiBlur
canonical_url: "https://www.modelscope.cn/models/LiblibAI/FLUX.1-dev-LoRA-AntiBlur"
md_url: "https://www.modelscope.cn/models/LiblibAI/FLUX.1-dev-LoRA-AntiBlur.md"
repository: LiblibAI/FLUX.1-dev-LoRA-AntiBlur
last_updated: 2024-11-21
license: other
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
base_model:
  - black-forest-labs/FLUX.1-dev
base_model_relation: adapter
parameters: 343.7M
tensor_type:
  - F16
library_name:
  - pytorch
  - lora
  - diffusers
  - safetensors
frameworks:
  - pytorch
downloads: 2509
stars: 7
tags:
  - text-to-image
  - stable-diffusion
  - lora
  - diffusers
  - image-generation
  - flux
  - safetensors
---

# FLUX.1-dev-LoRA-AntiBlur

> FLUX.1-dev-LoRA-AntiBlur - LiblibAI 在 ModelScope 开源的模型。FLUX.1-dev-LoRA-AntiBlur

LiblibAI/FLUX.1-dev-LoRA-AntiBlur 是 ModelScope 魔搭社区上的 343.7M 参数text-to-image-synthesis模型，采用 other 许可，基于 black-forest-labs/FLUX.1-dev 构建。

- **Repository**: LiblibAI/FLUX.1-dev-LoRA-AntiBlur
- **License**: other
- **Tasks**: text-to-image-synthesis
- **Parameters**: 343.7M
- **Base model**: black-forest-labs/FLUX.1-dev
- **Tags**: text-to-image, stable-diffusion, lora, diffusers, image-generation, flux, safetensors
- **Downloads**: 2509
- **Stars**: 7
- **Last updated**: 2024-11-21

Source: https://www.modelscope.cn/models/LiblibAI/FLUX.1-dev-LoRA-AntiBlur

---

# FLUX.1-dev-LoRA-AntiBlur

This is a functional LoRA trained on FLUX.1-dev for deep DoF (Anti-Blur🔥) by [Vadim_Fedenko](https://www.shakker.ai/userpage/1f90018d803d4045b8dec4d627915098/publish) on [Shakker AI](https://www.shakker.ai/modelinfo/5c3fa3f1d5034e63be325196eae0b4f6?from=search). 
It may not be fancy, but it works.

<div class="container">
  <img src="./poster.jpg" width="1024"/>
</div>

<!-- ## Showcases
<Gallery /> -->

## Comparison

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

<div class="container">
  <img src="./compare1.png" width="1024"/>
</div>

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.

<div class="container">
  <img src="./compare2.png" width="1024"/>
</div>

## Trigger words
The trigger word is not required. The recommended scale is `1.0` to `1.5` in diffusers.

## Inference

```python
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](https://www.shakker.ai/modelinfo/5c3fa3f1d5034e63be325196eae0b4f6?from=search), where we provide an online interface to generate images.


## Acknowledgements
This model is trained by our copyrighted users [Vadim_Fedenko](https://www.shakker.ai/userpage/1f90018d803d4045b8dec4d627915098/publish). We release this model under permissions. The model follows [flux-1-dev-non-commercial-license](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md).