---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: 'Xmas 3D, A close-up shot of a wooden Santa Claus figure with a white beard and a red and white Santa hat. The figure is positioned on a white snow-covered object. To the left of the figure is a green Christmas tree with white dots on it. The background is blurred.'
  output:
    url: images/4.png
- text: 'Xmas 3D, A gingerbread house with intricate icing details stands on a white snowy base. The roof is decorated with colorful gumdrops and candy canes, and the windows glow with warm light from inside. Beside the house, there is a tiny chocolate snowman and a cluster of peppermint sticks. In the background, blurry Christmas decorations add depth to the scene.'
  output:
    url: images/5.png
- text: 'Xmas 3D, A detailed shot of a reindeer figurine standing on a snowy hill. The reindeer has golden antlers and is adorned with a red velvet harness decorated with silver bells. Around its hooves, there are tiny snowdrift details. In the background, you can see blurred silhouettes of more reindeer and pine trees lit by a soft moonlight glow.'
  output:
    url: images/6.png
- text: 'Xmas 3D, two gingerbread men are standing on a dark wooden table. The gingerbread man on the right is wearing a blue belt, a blue hat, and a blue ribbon around his waist. His eyes are glowing blue. His hands are raised in front of him, and his mouth is slightly open. He is holding a candy cane in his left hand. The candy cane is in the shape of a heart, with a white rope wrapped around it. In front of the ginger bread man, there is a white candy cane with a red number 2 on it. There are two red candy barks on the table, adding a pop of color to the scene. In the background, a row of gingerbread trees are adorned with green and white lights.'
  output:
    url: images/1.png
- text: 'Xmas 3D, a 3-D image of a Santa Claus hand holding a small brown bottle with a green ribbon tied around it. The bottle is tied with a brown ribbon and a green leaf. The background is a vibrant red, adding a pop of color to the scene. The hand holding the bottle is white, with a white Santa hat on the right side of the image.'
  output:
    url: images/2.png
- text: 'Xmas 3D, A close-up shot of a plastic snowman figurine. The snowman is facing the left side of the frame, with a smile on its face. It has a red hat with a pom-pom on the top of it, and a red scarf around its neck. There are two black circles on the bottom of the snowmans face. The background is blurred, and there are other objects in the background.'
  output:
    url: images/3.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Xmas 3D
license: creativeml-openrail-m
---
![asfdvsdvc.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/3WTcfv5fiKimGNjx4FXs9.png)

<Gallery />

# Model description for Flux-Xmas-3D-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           | 24 & 3100 |
| Epoch                     | 20   | Save Every N Epochs       | 1     |

    Labeling: florence2-en(natural language & English)
    
    Total Images Used for Training : 14 [ 3D Artstation ]
## 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-Xmas-3D-LoRA"
trigger_word = "Xmas 3D"  
pipe.load_lora_weights(lora_repo)

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

You should use `Xmas 3D` to trigger the image generation.

## Download model

Weights for this model are available in Safetensors format.

[Download](/strangerzonehf/Flux-Xmas-3D-LoRA/tree/main) them in the Files & versions tab.