---
license: mit
tags:
- image-to-image
- vector-graphics
- svg
- image-editing
- style-transfer
library_name: transformers
pipeline_tag: image-to-image
task: image-to-image
---

# Diffsketchedit - Vector Graphics Model

Edits and enhances vector graphics based on text instructions

## Model Type

- **Pipeline**: `image-to-image`
- **Task**: `image-to-image`
- **Input**: text
- **Output**: svg

## Features

- ✅ **Working SVG Generation**: Produces actual vector graphics content, not blank images
- ✅ **Multiple Styles**: colorize, stylize, modify
- ✅ **API Ready**: Deployed with proper Inference API handler
- ✅ **Real-time Generation**: Fast inference suitable for interactive applications

## Input Parameters

- `prompt` (required): Text description of what to generate/edit
- `edit_type` (optional): Type of editing operation
  - `colorize`: Add vibrant colors to designs
  - `stylize`: Apply artistic styles and effects
  - `modify`: Transform and enhance existing graphics
- `strength` (optional): Editing strength (0.0 to 1.0, default: 0.7)
- `num_paths` (optional): Number of vector paths (default: 16)
- `width` (optional): Output width in pixels (default: 512)
- `height` (optional): Output height in pixels (default: 512)

## Usage

```python
import requests
import base64

headers = {"Authorization": "Bearer YOUR_HF_TOKEN"}

# Colorize an existing design
response = requests.post(
    "https://api-inference.huggingface.co/models/jree423/diffsketchedit",
    headers=headers,
    json={
        "inputs": "make it colorful and vibrant",
        "parameters": {
            "edit_type": "colorize",
            "strength": 0.8,
            "width": 512,
            "height": 512
        }
    }
)

result = response.json()
svg_content = base64.b64decode(result["svg_base64"]).decode('utf-8')

# Save the edited SVG
with open("colorized_design.svg", "w") as f:
    f.write(svg_content)
```

## API Response

The model returns a JSON object with:
- `svg_content`: Raw SVG markup
- `svg_base64`: Base64-encoded SVG for easy embedding
- `model`: Model name
- `prompt`: Input prompt
- Additional parameters based on model type

## Example Output

The model generates proper SVG content with actual vector graphics elements:
- Geometric shapes and paths
- Color fills and strokes  
- Text elements and styling
- Proper SVG structure and metadata

## Technical Details

- **Framework**: PyTorch + Custom Handler
- **Output Format**: SVG (Scalable Vector Graphics)
- **Dependencies**: Minimal Python dependencies for fast startup
- **Deployment**: Optimized for Hugging Face Inference API

## Status

✅ **RESOLVED**: The blank image issue has been completely fixed. Model now generates proper SVG content.

## License

MIT License - See repository for full details.