---
tags:
  - text-to-image
  - lora
  - diffusers
  - template:diffusion-lora
widget:
- output:
    url: images/111.png
  text: 'Glass Cube, a clear cube-shaped glass block is adorned with a chrome-plated metal bar nailed to the side of it. The bar is positioned in the center of the block, with the words Na Sodium 11 written below it in a bold black font. The background is a stark white, creating a stark contrast to the cube.'
- output:
    url: images/222.png
  text: 'Glass Cube, a clear cube-shaped object is adorned with the words Li Lithium 3 in black letters. The cube is positioned against a stark white backdrop, creating a stark contrast to the object. The objects reflection is visible on its surface, adding a touch of depth to the image.'
- output:
    url: images/333.png
  text: 'Glass Cube, Captured from a high-angle perspective on a pristine white backdrop, a square cube-shaped object is adorned with two silver balls, one of which is labeled Ni nickel 28 and the other "Nickel 28". Each ball has a speckled texture, adding a touch of contrast to the composition. The cube-like object casts a shadow on the right side of the frame, adding depth to the image.'
- output:
    url: images/444.png
  text: 'Glass Cube, a black rock is encased in a clear cube-shaped plastic box. The rocks surface is smooth and glossy, with a rough texture and a smooth surface. Below the rock, the words C Carbon 6 are written in bold black letters. The background is a stark white, creating a stark contrast to the black rock.'
- output:
    url: images/555.png
  text: 'Glass Cube, A brilliant gold nugget is suspended within a crystal-clear cube-shaped block. The nugget has a naturally irregular surface that shimmers under the lighting, casting faint golden reflections on the inner glass walls. Below the nugget, the words Au Gold 79 are printed in bold black text. The background is stark white, allowing the rich, metallic hue of the gold to stand out vividly. A subtle shadow from the cube falls to the right, adding dimension to the composition.'
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Glass Cube
---

![1.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/9tZ4Pb9WViboSswJjBxjg.png)

# Flux-Material-Glass-Cube

<Gallery />

---

# Model description for Flux-Material-Glass-Cube

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           | 18 & 2100 |
| Epoch                     | 14  | Save Every N Epochs       | 1  |

    Labeling: florence2-en(natural language & English)
    
    Total Images Used for Training : 17

## 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-Material-Glass-Cube"
trigger_word = "Glass Cube"  
pipe.load_lora_weights(lora_repo)

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

## Trigger words

You should use `Glass Cube` to trigger the image generation.


## Download model


[Download](/strangerzonehf/Flux-Material-Glass-Cube/tree/main) them in the Files & versions tab.