Baike.dev
All toolsOnline toolsAI codingTrendingOpen sourceNewsSubmit
Log in
< Back to tools
L

libavif

> 编程语言
Open source

libavif - Library for encoding and decoding .avif files

2.1K stars0 likes0 views
WebsiteGitHub

About

libavif - Library for encoding and decoding .avif files

libavif

This library aims to be a friendly, portable C implementation of the AV1 Image File Format, as described here:

https://aomediacodec.github.io/av1-avif/

It can encode and decode all AV1 supported YUV formats and bit depths (with alpha). In addition to the library, encoder and decoder command line tools are also provided (avifenc and avifdec).

It is recommended that you check out/use tagged releases instead of just using the main branch. We will regularly create new versions as bug fixes and features are added.

Command line tool usage

bash
avifenc -q 75 input.[jpg|png|y4m] output.avif
avifdec output.avif decoded.png

See avifenc --help for all options.

API usage

Please see the examples in the "examples" directory. If you're already building libavif, enable the CMake option AVIF_BUILD_EXAMPLES in order to build and run the examples too.

Installation

libavif is a package in most major OSs.

Windows

bash
vcpkg install libavif

You can also download the official windows binaries on the release page.

macOS

Homebrew:

bash
brew install libavif

MacPorts:

bash
sudo port install libavif

Linux

Debian-based distributions:

bash
sudo apt install libavif-dev

Red Hat-based distributions:

bash
sudo yum -y install libavif

MinGW

For the "default" MSYS2 UCRT64 environment:

bash
pacman -S mingw-w64-ucrt-x86_64-libavif

Build Notes

Building libavif requires CMake. See Build Command Lines below for example command lines.

Controlling Dependencies

CMake flags like AVIF_CODEC_AOM, AVIF_LIBYUV, etc. allow enabling or disabling dependencies. They can take three possible values:

  • OFF: the dependency is disabled.
  • SYSTEM: the dependency is expected to be installed on the system.
  • LOCAL: the dependency is built locally. In most cases, CMake can automatically download and build it. For some dependencies, you need to run the associated script in the ext/ subdirectory yourself. In cases where CMake handles downloading the dependency, you can still call the script in ext/ if you want to use a different version of the dependency (e.g. by modifying the script) or make custom code changes to it. If a directory with the dependency exists in the ext/ directory, CMake will use it instead of downloading a new copy.

Codec Dependencies

No AV1 codecs are enabled by default. You should enable at least one of them by setting any of the following CMake options to LOCAL or SYSTEM, depending on whether you want to use a locally built or a system installed version (e.g. -DAVIF_CODEC_AOM=LOCAL):

  • AVIF_CODEC_AOM for libaom (encoder and decoder)
  • AVIF_CODEC_DAV1D for dav1d (decoder)
  • AVIF_CODEC_LIBGAV1 for libgav1 (decoder)
  • AVIF_CODEC_RAV1E for rav1e (encoder)
  • AVIF_CODEC_SVT for SVT-AV1 (encoder)

When set to SYSTEM, these libraries (in their C API form) must be externally available (discoverable via CMake's FIND_LIBRARY) to use them, or if libavif is a child CMake project, the appropriate CMake target must already exist by the time libavif's CMake scripts are executed.

Libyuv Dependency

Libyuv is an optional but strongly recommended dependency that speeds up color space conversions. It's enabled by default with a value of SYSTEM, so it's expected to be installed on the system. It can either be built locally instead by using -DAVIF_LIBYUV=LOCAL or disabled with -DAVIF_LIBYUV=OFF.

Tests

A few tests written in C can be built by enabling the AVIF_BUILD_TESTS CMake option.

The remaining tests require GoogleTest, and can be built by enabling AVIF_BUILD_TESTS and setting AVIF_GTEST to SYSTEM or LOCAL.

Additionally, fuzzing tests require fuzztest, see also fuzzing test instructions in ext/oss-fuzz/README.md.

Code coverage is available by enabling AVIF_ENABLE_COVERAGE then building the avif_coverage target, e.g. make avif_coverage -j. It requires compiling with clang (-DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++) and LLVM must be installed on the system.

Build Command Lines

The following instructions can be used to build the libavif library and the avifenc and avifdec tools.

Build using installed dependencies

To link against the already installed aom, libjpeg, libpng and libyuv dependency libraries (recommended):

bash
git clone -b v1.2.1 https://github.com/AOMediaCodec/libavif.git
cmake -S libavif -B libavif/build -DAVIF_CODEC_AOM=SYSTEM -DAVIF_BUILD_APPS=ON
cmake --build libavif/build --config Release --parallel

Build everything from scratch

For development and debugging purposes:

bash
git clone -b v1.2.1 https://github.com/AOMediaCodec/libavif.git
cmake -S libavif -B libavif/build -DCMAKE_BUILD_TYPE=Debug -DBUILD_SHARED_LIBS=OFF -DAVIF_CODEC_AOM=LOCAL -DAVIF_LIBYUV=LOCAL -DAVIF_LIBSHARPYUV=LOCAL -DAVIF_JPEG=LOCAL -DAVIF_ZLIBPNG=LOCAL -DAVIF_BUILD_APPS=ON
cmake --build libavif/build --config Debug --parallel

Prebuilt Binaries (Windows)

Statically-linked avifenc.exe and avifdec.exe can be downloaded from the Releases page.

Development Notes

Please check the wiki for extra resources on libavif, such as the Release Checklist.

The libavif library is written in C99. Most of the tests are written in C++14.

Formatting

Use clang-format to format the sources from the top-level folder (clang-format-19 preferred):

bash
clang-format -style=file -i \
  apps/*.c apps/*/*.c apps/*/*.cc apps/*/*.h examples/*.c \
  include/avif/*.h src/*.c src/*.cc \
  tests/*.c tests/*/*.cc tests/*/*.h

Use cmake-format to format the CMakeLists.txt files from the top-level folder:

bash
cmake-format -i \
  CMakeLists.txt \
  tests/CMakeLists.txt \
  cmake/Modules/*.cmake \
  contrib/CMakeLists.txt \
  contrib/gdk-pixbuf/CMakeLists.txt \
  android_jni/avifandroidjni/src/main/jni/CMakeLists.txt

License

Released under the BSD License.

…

Issues· 0 open

View all issuesOpen on GitHub

No open issues yet, or sync has not completed.

> Tags

C

No comments yet. Be the first to share.

> Details

PublishedAug 1, 2026
UpdatedSep 17, 2026
Category编程语言
PricingOpen source

> Related tools

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言