---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- output:
    url: images/111.png
  text: 'aura lights, a vibrant black smiley face is adorned with a glowing circle of light. The face is encircled by a pair of shiny, shiny hands, adding a touch of warmth to the scene. The background is a solid black, creating a stark contrast to the black figure. The figures arms and hands are positioned in a way that creates a sense of balance, adding depth to the composition.'
- output:
    url: images/222.png
  text: 'aura lights, a close-up shot of a womans face, adorned with a white mask. The womans head is adorned with curly hair, a white rose, and a black scarf. The background is a solid black, creating a stark contrast to the womans figure. The figures face is covered in a black mask, adding a pop of color to the scene. The scarf is draped over her shoulders, adding depth to the composition.'
- output:
    url: images/333.png
  text: 'aura lights, a close-up of a statue of a womans face is adorned with a white robe adorned with black stripes. The womans head is adorned in a headdress adorned with white flowers, adding a touch of color to the scene. The background is dark, creating a stark contrast to the statue.'
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: aura lights
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
---

![aura.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/m0d1zgJrx2NNfj2rLnHZd.png)

# Aura-Lights

<Gallery />

---

# Model description for Aura-Lights

Image Processing Parameters 

| Parameter                 | Value  | Parameter                 | Value  |
|---------------------------|--------|---------------------------|--------|
| LR Scheduler              | constant | Noise Offset              | 0.03   |
| Optimizer                 | AdamW  | Multires Noise Discount   | 0.1    |
| Network Dim               | 64     | Multires Noise Iterations | 10     |
| Network Alpha             | 32     | Repeat & Steps           | 20 & 2300 |
| Epoch                     | 15  | Save Every N Epochs       | 1  |

    Labeling: florence2-en(natural language & English)
    
    Total Images Used for Training : 20

## Best Dimensions & Inference

| **Dimensions** | **Aspect Ratio** | **Recommendation**       |
|-----------------|------------------|---------------------------|
| 1280 x 832      | 3:2              | Best                     |
| 1024 x 1024     | 1:1              | Default                  |

### Inference Range

- **Recommended Inference Steps:** 30-35

## Setting Up
```python
import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "strangerzonehf/Aura-Lights"
trigger_word = "aura lights"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)
```

## Trigger words

You should use `aura lights` to trigger the image generation.

## Download model


[Download](/strangerzonehf/Aura-Lights/tree/main) them in the Files & versions tab.