---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
  - text: 'Realism H6, a young woman with dark brown hair and brown eyes stands against a stark white backdrop. She is dressed in a bright yellow polka dotted dress, adorned with blue polka dots. Her hair is pulled back, cascading down her shoulders, framing her face. Her lips, painted a vibrant shade of pink, is adorned with tiny black dots, adding a touch of contrast to her yellow dress.'
    output:
      url: images/1111.png

  - text: 'Realism H6, Woman in a red jacket, snowy, in the style of hyper-realistic portraiture, caninecore, mountainous vistas, timeless beauty, palewave, iconic, distinctive noses --ar 72:101 --stylize 750 --v 6'
    output:
      url: images/2222.png

  - text: 'Realism H6, Captured at eye-level, a close-up portrait of a woman with long brown hair, a black sweater with white polka dots on it, is positioned in front of a blue wall. The womans face is angled slightly to the right, her lips are slightly parted, and her eyes are slightly open. Her hair cascades over her shoulders, adding a touch of warmth to her face.'
    output:
      url: images/3333.png

  - text: 'Realism H6, a close-up shot of a mans face features a neutral expression. The mans eyes are a piercing blue, and his hair is a rich brown, with a few wispy bangs. His eyebrows are a darker shade of brown, adding a touch of texture to his face. He is dressed in a black blazer, a black t-shirt, and a black collared shirt. The backdrop is a stark white, creating a stark contrast to the mans attire.'
    output:
      url: images/4444.png

base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Realism H6
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
---
![2.png](https://cdn-uploads.huggingface.co/production/uploads/65bb837dbfb878f46c77de4c/X4Aecst7dVTHvQzIeJIC7.png)

# Realism-H6-Flux

<Gallery />

# Model description for Realism-H6-Flux

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           | 23 & 3990 |
| Epoch                     | 33   | Save Every N Epochs       | 1    |

    Labeling: florence2-en(natural language & English)
  
    Total Images Used for Training : 37
    
## 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/Realism-H6-Flux"
trigger_word = "Realism H6"  
pipe.load_lora_weights(lora_repo)

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

You should use `Realism H6` to trigger the image generation.

## Download model

Weights for this model are available in Safetensors format.

[Download](/strangerzonehf/Realism-H6-Flux/tree/main) them in the Files & versions tab.