Text-to-Image
Diffusers
lora
template:diffusion-lora
FallenArt-Flux / README.md
prithivMLmods's picture
Update README.md
955d564 verified
---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: 'Fallen Art, a medium-sized portrait of a mans face is displayed against a light blue background. The mans head is angled slightly to the left, and his eyes are squinted. His hair is cut in a bob, adding a pop of color to the scene. His eyes are a piercing blue, while his eyebrows are a darker shade of brown. His lips are a lighter shade of blue, and he is wearing a white button-down shirt with a black collar. His neck is adorned with a small cross pendant.'
output:
url: images/111.png
- text: 'Fallen Art, Captured from a low-angle perspective on a vibrant purple backdrop, a portrait of a mans face is depicted in a stark c
ontrast of purple and white. The mans head is adorned with a black wig, adding a pop of color to the scene. His eyes are a piercing blue, while his lips are a lighter shade of white. His eyebrows are a darker shade of black, adding depth to the composition. His jacket is a deep purple, with a white collar and a white shirt tucked into the jacket. His hands are clasped in front of his face, adding texture to the overall composition.'
output:
url: images/222.png
- text: 'Fallen Art, Captured from a low-angle perspective on a vibrant pink backdrop, a womans face is adorned with a pink heart-shaped necklace. Her hair is styled in a ponytail, adding a pop of color to her face. Her eyes are a piercing blue, while her lips are a lighter shade of pink. Her eyebrows are a darker shade of brown, and her mouth is slightly open, as if she iss smiling. Her lips are slightly parted, and she has a slight smile on her lips. The womans shadow is cast on the left side of the image, adding depth to the composition.'
output:
url: images/333.png
- text: 'Fallen Art, from a low front-facing angle, a womans head tilts slightly back, her face painted in soft whites and grays. Her eyes are closed, lips pursed in subtle thought. Her jet-black hair is styled in waves, adorned with silver pins shaped like stars. The backdrop is a faded crimson, and her silhouette is rim-lit, casting a soft halo effect. A delicate black lace glove covers her hand as it brushes her chin.'
output:
url: images/444.png
- text: 'Fallen Art, a close-up portrait of a womans face captured from a side angle, her gaze directed off-frame to the right. Her skin is painted in soft peach hues, contrasted by a teal backdrop. Her hair is cut in sharp geometric bangs, dyed in a gradient from white to pale green. Her lips are a deep magenta, and a single silver teardrop earring dangles from her left ear. Her collarbone is visible, adorned with a thin chain necklace holding a tiny lock.'
output:
url: images/555.png
- text: 'Fallen Art, a centered portrait of a young woman with a bold, confident expression. Her background is a vibrant neon pink, and her long hair is tied into two high ponytails, dyed electric blue. Her eyes are outlined in black liner, exaggerated to a dramatic wing. She wears a transparent raincoat over a silver tank top, with layered metallic necklaces across her neck. Her lip color is matte black, adding a punch to her presence.'
output:
url: images/666.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Fallen Art
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
---
![DFG.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/n0hpkIfPmboKIsFvbjqDf.png)
# FallenArt-Flux
<Gallery />
# Model description for FallenArt-Flux
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 | 25 & 3450 |
| Epoch | 23 | Save Every N Epochs | 1 |
Labeling: florence2-en(natural language & English)
Total Images Used for Training : 26
## 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/FallenArt-Flux"
trigger_word = "Fallen Art"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
```
## Trigger words
You should use `Fallen Art` to trigger the image generation.
## Download model
Weights for this model are available in Safetensors format.
[Download](/strangerzonehf/FallenArt-Flux/tree/main) them in the Files & versions tab.