Open-source licences

The video tools, and the audio tools such as WAV to MP3, are built on open-source software. Nothing you process is uploaded: these components run inside your browser. An MP3 is written by the LAME encoder, through @mediabunny/mp3-encoder below or, for some files, the ffmpeg.wasm core.

• ffmpeg.wasm core (@ffmpeg/core): GPL-2.0-or-later (full text of the GPL-2.0), because it bundles x264. It is used only for some files. Source: github.com/ffmpegwasm/ffmpeg.wasm, tag v0.12.10 (core version 0.12.10). The wrapper @ffmpeg/ffmpeg and @ffmpeg/util are MIT.
• Mediabunny 1.55.5: MPL-2.0. Source: github.com/Vanilagy/mediabunny, tag v1.55.5.
• @mediabunny/mp3-encoder 1.61.4: MPL-2.0. It contains a WebAssembly build of the LAME MP3 encoder (lame.sourceforge.io), version 3.100, which is LGPL-2.0-or-later. Source of the package: github.com/Vanilagy/mediabunny, packages/mp3-encoder, tag v1.61.4. Source of LAME 3.100: sourceforge.net/projects/lame, version 3.100.
• @mediabunny/aac-encoder 1.61.4: MPL-2.0. It contains a WebAssembly build of FFmpeg’s AAC encoder, which is LGPL-2.1-or-later. Source of the package: github.com/Vanilagy/mediabunny, packages/aac-encoder, tag v1.61.4. Source of FFmpeg: ffmpeg.org/download.html.
• client-zip: MIT.
• gifenc 1.0.3: MIT. It writes the animated GIFs.
• Inter (Bold): SIL Open Font License 1.1, by Rasmus Andersson. Source: github.com/rsms/inter. It is drawn into your videos by the text and subtitle tools. The licence text is served beside the font: OFL.txt.

The image tools are built on open-source software. Nothing you process is uploaded: these components run inside your browser.

• pica 10.0.3 (with glur and multimath): MIT. It resizes pictures. Source: github.com/nodeca/pica.
• exifr 7.1.3: MIT. It reads the details stored in a photo (EXIF, GPS, XMP, IPTC). Source: github.com/MikeKovarik/exifr.
• pdf-lib 1.17.1 (with @pdf-lib/standard-fonts and @pdf-lib/upng): MIT. It writes PDFs. It brings pako (MIT and Zlib) and tslib (0BSD). Source: github.com/Hopding/pdf-lib.
• gifenc 1.0.3: MIT, as above. It writes GIFs here too.
• gifuct-js 2.1.2 (with js-binary-schema-parser): MIT. It reads the frames of a GIF. Source: github.com/matt-way/gifuct-js.
• UTIF (utif) 3.1.0: MIT. It reads and writes TIFF. Source: github.com/photopea/UTIF.js.
• imagetracerjs 1.2.6: the Unlicense (public domain). It traces a picture into an SVG. Source: github.com/jankovicsandras/imagetracerjs.
• wasm-feature-detect 1.9.0: Apache-2.0. Source: github.com/GoogleChromeLabs/wasm-feature-detect.
• jSquash (@jsquash/jpeg 1.6.0, @jsquash/webp 1.5.0, @jsquash/avif 2.1.1, @jsquash/oxipng 2.3.0): Apache-2.0 (full text of the Apache-2.0). Source: github.com/jamsinclair/jSquash. They load WebAssembly builds of these encoders, each only when a picture is saved in its format:
  • MozJPEG (built on libjpeg-turbo): the IJG licence, the Modified (3-clause) BSD licence and the zlib licence (licence text). This software is based in part on the work of the Independent JPEG Group. Source: github.com/mozilla/mozjpeg.
  • libwebp: BSD-3-Clause (licence text). Source: chromium.googlesource.com/webm/libwebp. It also encodes each frame of an animated WebP (GIF to WebP); our own code puts the frames together.
  • libavif 1.0.1: BSD-2-Clause (licence text), with libaom: BSD-2-Clause (licence text) and the Alliance for Open Media Patent License 1.0 (licence text). Sources: github.com/AOMediaCodec/libavif, tag v1.0.1 and aomedia.googlesource.com/aom.
  • oxipng: MIT (licence text). Source: github.com/shssoichiro/oxipng.
• libheif and libde265 (through libheif-js 1.23.5) are licensed under the GNU LGPL 3.0 (full text of the LGPL-3.0, which builds on the GPL-3.0). They are loaded as separate, unmodified files from /image-engine/libheif-js-1.23.5/ only when you open a HEIC picture that your browser cannot read itself. Their source code is at github.com/catdad-experiments/libheif-js, tag v1.23.5, github.com/strukturag/libheif and github.com/strukturag/libde265. You may replace these files with your own build. The folder’s SOURCE.txt lists the files, their checksums and how they are loaded.
• Inter (Bold): SIL Open Font License 1.1, as above. Anton (Regular), by the Anton Project Authors: SIL Open Font License 1.1 (licence text). Source: github.com/googlefonts/AntonFont. The tools that put words on a picture use them.

Two of the sample pictures behind “Try it with a sample” are photos from Pexels, used under the Pexels licence. They were made smaller and their details (such as camera and place) were removed. The other samples are drawn by us.

• Portrait (portrait.jpg): photo by cottonbro studio. Source: pexels.com/photo/a-woman-in-beige-top-10209460.
• Café (cafe.jpg): photo by Ánh Đặng. Source: pexels.com/photo/cozy-urban-cafe-interior-with-warm-lighting-34094296.

The pages you are reading are built from the parts below, which also reach your browser.

• Next.js, React and React DOM: MIT.
• Lucide (lucide-react, the icons): ISC.
• with icons (the pictures on the image and video tool cards), by the with icons contributors: MIT. Source: github.com/withevergrow/withicons. The licence text is served here: MIT-withicons.txt.
• Archivo (the site font), by the Archivo Project Authors: SIL Open Font License 1.1. Source: github.com/Omnibus-Type/Archivo. The licence text is served here: Archivo-OFL.txt.

Our server also uses GeoLite2 Country, a list that matches internet addresses to countries, to tell whether you are in Europe and so whether to ask about cookies first (see the privacy page). It never reaches your browser. This product includes GeoLite Data created by MaxMind, available from maxmind.com, under the GeoLite End User License Agreement.

For Remove background and Add light to image, our server works on a small copy of your photo (see the privacy page) with two models made by others. Neither reaches your browser. MODNet, which finds the person, in the published weights converted by Xenova: Apache-2.0 (full text of the Apache-2.0). Source: github.com/ZHKKKe/MODNet. Depth Anything V2 Small, for Add light to image only: Apache-2.0. Source: github.com/DepthAnything/Depth-Anything-V2.

Back to the video tools