---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: 'Super Detail, A close-up shot of a man with a brown hat on his head. His eyes are blue and he has brown hair. His hair is wet from the rain. The background is blurred.'
  output:
    url: images/SR1.png
- text: 'Super Detail, A close-up shot of a womans face, taken from a low-angle perspective. The womans eyes are a piercing of turquoise, and her hair is a vibrant shade of brown. Her lips are painted a deep red, with a slight smile. Her eyebrows are a light brown, adding a touch of texture to her face. The background is dark, creating a stark contrast to the womans skin.'
  output:
    url: images/SR3.png
- text: 'Super Detail, a close-up shot of a womans head and shoulders is seen against a vibrant red backdrop. The womans face is adorned with a white face, adorned with blue eyes, and her brown hair cascades over her shoulders. She is wearing a red turtleneck, with a ribbed collar. Her lips are painted a vibrant shade of red, adding a pop of color to her face. Her eyebrows are a darker shade of blue, adding depth to the composition.'
  output:
    url: images/SR4.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Super Detail
license: creativeml-openrail-m
---
# Flux-Super-Detail-LoRA
<span style="color:red">The model is still in the training phase. This is not the final version and may contain artifacts and perform poorly in some cases.</span>

<Gallery />

**prithivMLmods/Flux-Super-Detail-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           | 15 & 2470 |
| Epoch                     | 10    | Save Every N Epochs       | 1     |

    Labeling: florence2-en(natural language & English)
    
    Total Images Used for Training : 15

## Best Dimensions

- 768 x 1024 (Best)
- 1024 x 1024 (Default)

## 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 = "prithivMLmods/Flux-Super-Detail-LoRA"
trigger_word = "Super Detail"  
pipe.load_lora_weights(lora_repo)

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

You should use `Super Detail` to trigger the image generation.

## Download model

Weights for this model are available in Safetensors format.

[Download](/prithivMLmods/Flux-Super-Detail-LoRA/tree/main) them in the Files & versions tab.