README.md

---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
- '3:2'
- NFT
- Art
- Design
widget:
- text: 'NFT V4, Captured at eye-level, a close-up shot of a cats face is depicted. The cats fur is a vibrant shade of pink, with a mix of brown and yellow. The eyes are tinted black, with yellow accents. The ears of the cat are a light pink, and the cats nose is a darker pink. The background is a dark brown, adding a pop of color to the scene.'
  output:
    url: images/1.png
- text: 'NFT V4, a close-up shot of a spiderman cat wearing a red and black spiderman costume. The cats face is adorned with a black nose, black ears, and white eyes. The background is a vibrant shade of pink, adding a pop of color to the scene.'
  output:
    url: images/2.png
- text: 'NFT V4, Captured in a low-angle perspective, a medium-sized cats face is adorned with a maroon baseball cap, white sunglasses, and a silver chain around its ears. The cats eyes are tinted black, and the cats whiskers are white, adding a pop of color to the scene. Its mouth is slightly open, as if its lips are pursed. Its nose is a dark brown, and its mouth is a lighter shade of brown, while the cat is wearing a white t-shirt with a silver necklace around its neck. The background is a solid black, creating a stark contrast to the cat.'
  output:
    url: images/3.png
- text: 'NFT V4, a medium-sized bulldog is adorned with a black leather jacket and a silver chain around its neck. The bulldogs eyes are covered in a pair of tinted sunglasses, adding a pop of color to the otherwise monochromatic image. The dogs mouth is slightly open, revealing a small white cigarette in its mouth. The background is a deep blue, creating a stark contrast to the dogs face.'
  output:
    url: images/4.png
- text: 'NFT V4: A head-on close-up of a fluffy rabbits face with multicolored fur in shades of teal, purple, and white. The rabbit is wearing round gold-rimmed glasses and a black bowtie. Its nose is pale pink, and its whiskers are long and white. The background is a gradient of lavender and light yellow, adding a soft and dreamy feel to the scene.'
  output:
    url: images/5.png
- text: 'NFT V4: A raccoon depicted in a playful pose, wearing a green hoodie with a "Game Over" text on the front. The raccoons fur is a mix of brown and gray with white accents on its face. Its nose is black, and it has mischievous green eyes. The background is a pixelated video game landscape with floating clouds and coins.'
  output:
    url: images/6.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: NFT V4
license: creativeml-openrail-m
---
![strangerzonehf/Flux-NTFv4-Designs-LoRA](images/xx.png)

<Gallery />

## Model description for NTFv4 Designs 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           | 25 & 3170 |
| Epoch                     | 16   | 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-NTFv4-Designs-LoRA"
trigger_word = "NFT V4"  
pipe.load_lora_weights(lora_repo)

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

## Trigger words

You should use `NFT V4` to trigger the image generation.

## Download model

Weights for this model are available in Safetensors format.

[Download](/strangerzonehf/Flux-NTFv4-Designs-LoRA/tree/main) them in the Files & versions tab.