Open-Source Licenses

Our browser-based AI tools are built on open-source models and libraries. We are grateful to the research community. This page lists the licenses for all third-party software we use.

AI Models

BEN2 — Background Erase Network

Used by: Background Remover

• Author: Prama LLC
• License: MIT
• Source: huggingface.co/PramaLLC/BEN2
• ONNX Conversion: onnx-community/BEN2-ONNX (MIT)
• Paper: arXiv:2501.06230

MobileSAM — Mobile Segment Anything Model

Used by: Smart Segmentation

• Author: Chaoning Zhang et al.
• License: Apache 2.0
• Source: github.com/ChaoningZhang/MobileSAM

UltraFace-RFB-320

Used by: Face Detection & Blurring

• Author: Ultra-Light-Fast-Generic-Face-Detector-1MB contributors
• License: MIT
• Source: github.com/Linzaer/Ultra-Light-Fast-Generic-Face-Detector-1MB

Libraries & Frameworks

Transformers.js

Used by: Background Remover

• Author: Hugging Face
• License: Apache 2.0
• Source: github.com/huggingface/transformers.js

ONNX Runtime Web

Used by: Background Remover, Smart Segmentation, Face Detection & Blurring

• Author: Microsoft
• License: MIT
• Source: github.com/microsoft/onnxruntime

wasm-bindgen

Used by: Pixelize, Palletize

• Author: The Rust and WebAssembly Working Group
• License: Apache 2.0 / MIT
• Source: github.com/rustwasm/wasm-bindgen

Marked

Used by: Markdown Editor

• Author: MarkedJS contributors
• License: MIT
• Source: github.com/markedjs/marked

Turndown

Used by: Markdown Editor

• Author: Dom Christie
• License: MIT
• Source: github.com/mixmark-io/turndown

QRCode.js

Used by: QR Code Generator

• Author: Kazuhiko Arase
• License: MIT
• Source: github.com/soldair/node-qrcode

---

All models and libraries listed above are licensed under MIT or Apache 2.0, permitting commercial use. No CC BY-NC or other restrictive-license models are used in production.