README.md

---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
- 3DXL
- Flux
- Part-0001
- 3D-Character-Design
widget:
- text: >-
    3DXLP1, a medium-sized woman stands in front of a dark gray background. She
    is dressed in a black bra, adorned with a scalloped design. Her hair is
    styled in a sleek bob, adding a touch of texture to her face. Her lips are
    painted a vibrant red, while her lips are pursed. Her right hand is raised
    in the air, holding a cigarette in her left hand. Her eyes are a piercing
    blue, and her eyebrows are a darker shade of brown. The lighting is subdued,
    creating a stark contrast against the dark background.
  output:
    url: images/4.png
- text: >-
    3DXLP1, a close-up shot of a bald man with blue eyes and a smile on his
    face. He is wearing a black and gray striped scarf around his neck, with a
    silver lanyard hanging from the collar. The backdrop is a dark gray,
    creating a stark contrast to the bald mans face.
  output:
    url: images/1.png
- text: >-
    3DXLP1, A close-up portrait of a man with long brown hair and a goatee. He
    is wearing a dark green button-down shirt. His eyes are blue and he has a
    nose ring on his nose. His eyebrows are covered in red marks. The background
    is a solid gray color.
  output:
    url: images/2.png
- text: >-
    3DXLP1, a close-up shot of a womans face features a neutral expression. Her
    eyes are a piercing blue, and her hair is a vibrant shade of brown. She is
    wearing a black turtleneck, with a yellow vest draped over her shoulders.
    The background is a solid black, creating a stark contrast to the womans
    outfit.
  output:
    url: images/3.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: 3DXLP1
license: creativeml-openrail-m
---
![Flux-3DXL-Partfile-0001](images/1x.png)

<Gallery />

## Model description for 3DXL Partfile 0001

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 & 3100 |
| Epoch                     | 20   | Save Every N Epochs       | 1     |

    Labeling: florence2-en(natural language & English)
    
    Total Images Used for Training : 27

## 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-3DXL-Partfile-0001"
trigger_word = "3DXLP1"  
pipe.load_lora_weights(lora_repo)

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

You should use `3DXLP1` to trigger the image generation.

## Download model

Weights for this model are available in Safetensors format.

[Download](/strangerzonehf/Flux-3DXL-Partfile-0001/tree/main) them in the Files & versions tab.