---
title: Text-to-Image Gradio Template
emoji: 🖼
colorFrom: purple
colorTo: red
sdk: gradio
sdk_version: 5.43.1
app_file: app.py
pinned: false
---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

# Stable Diffusion v1.5 Finetuned on Zappos Shoes (LoRA)

This is a finetuned version of Stable Diffusion v1.5, trained with LoRA on the Zappos Shoes Dataset.
The model is specialized in generating realistic shoe images such as sneakers, boots, and sandals.

## Model Description

- **Base model**: [Stable Diffusion v1.5](https://huggingface.co/runwayml/stable-diffusion-v1-5)
- **Finetuning method**: LoRA
- **Dataset**: [UT Zappos50K Shoes Dataset](http://vision.cs.utexas.edu/projects/finegrained/utzap50k/)
- **Primary use**: Generating realistic images of shoes
- **Optimal hardware**: GPU (for fast inference), but this demo runs on CPU in Hugging Face Spaces

## Example Usage

### Text-to-Image Prompt

```python
from huggingface_hub import InferenceClient

client = InferenceClient("benisonjac/stable-diffusion-finetune")

image = client.text_to_image("a photo of a high-top sneaker, futuristic design")
image.save("sneaker.png")
```

### Example Prompt
```
a photo of a high-top sneaker, futuristic design
```

## Running Locally

### Clone the repo
```bash
git clone https://huggingface.co/Shoe-Gen/Shoe_Generator_Hosting
cd shoe-generator
```

### Install dependencies
```bash
pip install -r requirements.txt
```

### Run the app
```bash
python app.py
```

### Access the demo
Open [http://localhost:7860](http://localhost:7860) in your browser.

## Demo (Hugging Face Space)

If you want to try it online, you can run the Space here:
👉 [Live Demo on Hugging Face Spaces](https://huggingface.co/spaces/Shoe-Gen/Shoe_Generator_Hosting)

## Limitations

- Works best for shoe-specific prompts (sneakers, sandals, boots).
- May not generalize well to unrelated objects.
- CPU inference is slow compared to GPU.

## License

- **Base model license**: [CreativeML Open RAIL-M](https://huggingface.co/spaces/CompVis/stable-diffusion-license)
- **Dataset**: [UT Zappos50K](http://vision.cs.utexas.edu/projects/finegrained/utzap50k/) (research only)
- This model is for research and educational purposes only.

## Acknowledgements

- [CompVis & RunwayML](https://huggingface.co/runwayml) for Stable Diffusion
- [UT Austin](http://vision.cs.utexas.edu/) for Zappos50K dataset
- Hugging Face for hosting