Open-Weight Model vs API Model#
15 examples. Quantitative metrics (MGE, MAE, Connectivity) require ground-truth alpha mattes. Use the view controls on each row to switch between transparency, chroma key, and alpha matte on all outputs. Metric scores are shown as ? until ground truth is added.


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Alexander Dummer on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Andreas Rasmussen on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Camila Quintero Franco on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Conor Samuel on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Ilja Tulit on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Ivan Dostal on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Jonny Clow on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Niko Tsviliov on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Vinicius Wiesehofer on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Carlo Sierra on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Lars Bo Nielsen on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Mayer Tawfik on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Photo by Norbert Braun on Unsplash


- MGE
- ?
- MAE
- ?
- Conn.
- ?
- MGE
- ?
- MAE
- ?
- Conn.
- ?
Methodology
- Alpha mattes are grayscale masks encoding per-pixel foreground probability. They capture fine edge detail (hair, fur, semi-transparent regions) that binary masks discard.
- Metrics are computed by comparing predicted mattes against a ground-truth reference: MGE measures edge sharpness, MAE measures pixel-level accuracy, and Connectivity penalises fragmented foreground regions. All three are lower-is-better.
- Scores are shown as ? when no ground-truth matte is available for that image. See the alpha matting evaluation benchmark for details on the evaluation methodology.
Photo Credits
Source images are courtesy of the photographers below via Unsplash.
Technical Specifications#
| Feature | Open-Weight Model | API Model (Cloud) |
|---|---|---|
| Status | Active (Open Source) | Active (Cloud API) |
| Availability | Apache 2.0 | API only |
| Pixel Accuracy | Excellent | Best Overall |
| Edge Quality | Excellent | Superior |
| Hardware | CPU / GPU | AWS Inferentia (dedicated AI chip) |
| Privacy | Images never leave your computer | Server-side processing (deleted after use) |
| Free Credits | Unlimited (run locally) | 50 credits on signup |
| Recommended Use | Privacy-sensitive, local processing | Highest quality requirements, production API |
Installation#
The Open-Weight Model runs locally; the cloud API is accessed via REST.
Open-Weight Model
Run instantly with Docker
docker run -p 80:80 withoutbg/app:latestPython
Install via pip
pip install withoutbgCloud API
API Request
Process images via the cloud API
curl -X POST https://api.withoutbg.com/v1/remove-background \
-H 'Authorization: Bearer YOUR_API_KEY' \
-F 'image=@input.jpg' \
-o output.pngSee the full API documentation for integration details.
Get Started with 50 Free Credits
Sign up and receive 50 free credits to test the cloud API. No credit card required.
- Create free account and get 50 credits instantly
- API documentation for integration