---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: 'Sketch Glow, An eye-level drawing of a woman with long, wavy blonde hair. The womans face is angled towards the left side of the image, with a slight smile on her lips. Her eyes are squinted, and her lips are pursed. Her hair cascades over her shoulders, framing her face. The background is a light purple, and the womans head is adorned with a bright orange glow, adding a pop of color to the scene.'
  output:
    url: images/4.png
- text: 'Sketch Glow, A delicate pencil sketch of a teenage girl with shoulder-length straight black hair. She is wearing a white blouse with a red bow at the collar. Her head is slightly tilted, and her eyes are wide open, giving her an innocent expression. She has small earrings shaped like stars. The background is a soft pastel yellow, adding warmth to the scene.'
  output:
    url: images/5.png
- text: 'Sketch Glow, A fine ink sketch of a small child sitting cross-legged on the ground, surrounded by scattered paper planes. The child is wearing a striped shirt and denim overalls, with a mischievous grin on their face. Their hair is short and slightly tousled. The background is a mix of light blue and white, with faint clouds sketched in for depth.'
  output:
    url: images/6.png
- text: 'Sketch Glow, A medium-sized sketch of a young Asian girl with long brown hair and bangs. Her eyes are squinted and her lips are pursed. Her hair is pulled back in a ponytail. She is wearing a light-colored blazer with a black collar. The blazer has two black lightning bolt designs on the front of it. The background is a pale pink color.'
  output:
    url: images/1.png
- text: 'Sketch Glow, A pencil sketch of a woman in a gray blazer with a green necklace around her neck. She is holding a wine glass in her right hand. Her left hand is holding the wine glass with her left hand. She has a cigarette in her mouth. Her eyes are open and she is wearing a black rimmed hat. Her hair is short and dark brown. Her lips are painted black. Her eyebrows are black and her hair is a light brown. The background is light gray.'
  output:
    url: images/2.png
- text: 'Sketch Glow, A medium-sized drawing of a boy is depicted on a white paper. The boys head is facing the left side of the image, and he is holding an open book in his hands. He is wearing a blue collared shirt, and a blue tie. His eyes are a piercing green, and his hair is a light brown color. He has a serious expression on his face. His hands are also holding the book, and the book is open.'
  output:
    url: images/3.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Sketch Glow
license: creativeml-openrail-m
---
![fdgvzdf.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/PoW7TwHYp3WDG9qGbKzHk.png)

<Gallery />

# Model description for Flux-Sketch-Glow-LoRA

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           | 21 & 2300 |
| Epoch                     | 15   | Save Every N Epochs       | 1     |

    Labeling: florence2-en(natural language & English)
    
    Total Images Used for Training : 16
    
## 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/Flux-Sketch-Glow-LoRA"
trigger_word = "Sketch Glow"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)
```
## Trigger words

You should use `Sketch Glow` to trigger the image generation.

## Download model

Weights for this model are available in Safetensors format.

[Download](/strangerzonehf/Flux-Sketch-Glow-LoRA/tree/main) them in the Files & versions tab.