# BiRefNet - Core.Today AI API

> State-of-the-art open-source background removal. BiRefNet's high-fidelity dichotomous image segmentation delivers excellent edge quality on hair, fur, and fine details.

- **Provider**: BiRefNet
- **Model ID**: men1scus/birefnet
- **Category**: Image Generation
- **Credits**: 5 per image
- **Speed**: Fast
- **Quality**: Ultra

## Features
- State-of-the-art segmentation quality
- Excellent edges on hair, fur, and fine details
- Adjustable processing resolution
- Simple one-input API
- Open-source (research-backed BiRefNet)

## Use Cases
- High-fidelity cutouts of people and animals
- Professional product photography
- Print-quality composites
- Difficult subjects (hair, lace, branches)
- Design asset preparation

## API Endpoint

Base URL: https://api.core.today/v1
Create Prediction: POST /predictions
Get Status: GET /predictions/{job_id}

## Authentication
Header: X-API-Key: YOUR_API_KEY

## Input Parameters

### Required
- **image**: string - Input image to remove the background from

### Optional
- **resolution**: string - Processing resolution in WxH format, e.g. '1024x1024'. Empty uses the default

## Examples

### High-Fidelity Portrait Cutout
Remove the background from a portrait with fine hair details
```json
{
  "model": "men1scus/birefnet",
  "input": {
    "image": "https://example.com/portrait.jpg"
  }
}
```

## Response Format
```json
{
  "job_id": "abc123",
  "status": "pending | processing | completed | failed",
  "result": "URL or data (when completed)"
}
```

## Usage Flow
1. POST /predictions with model and input -> receive job_id
2. GET /predictions/{job_id} -> poll until status is completed or failed
3. Result contains output URL(s)

## Tips
- Best-in-class for hair, fur, and semi-transparent edges
- Raise resolution (e.g. '2048x2048') for large source images when edges matter
- For bulk low-cost processing, Background Remover (2 credits) may be sufficient
- Output is a transparent PNG ready for compositing

## Documentation
https://replicate.com/men1scus/birefnet