Oxipng is a multithreaded lossless PNG/APNG compression optimizer. It can be used via a command-line interface or as a library in other Rust programs.

Oxipng for Windows can be downloaded from the Releases link on the GitHub page.

For MacOS or Linux, it is recommended to install from your distro's package repository, if possible. Oxipng is known to be packaged for the environments listed below.

Alternatively, oxipng can be installed from Cargo, via the following command:

cargo install oxipng

Oxipng can be built from source using the latest stable or nightly Rust. This is primarily useful for developing on oxipng.

git clone https://github.com/shssoichiro/oxipng.git
cd oxipng
cargo build --release
cp target/release/oxipng /usr/local/bin

The current minimum supported Rust version is 1.74.0.

Oxipng follows Semantic Versioning.

Oxipng is a command-line utility. An example usage, suitable for web, may be the following:

oxipng -o 4 --strip safe --alpha *.png

The most commonly used options are as follows:

• Optimization: -o 0 through -o 6 (or -o max), lower is faster, higher is better compression. The default (-o 2) is quite fast and provides good compression. Higher levels can be notably better but generally have increasingly diminishing returns.
• Strip: Used to remove metadata info from processed images. Used via --strip [safe,all]. Can save a few kilobytes if you don't need the metadata. "Safe" removes only metadata that will never affect rendering of the image. "All" removes all metadata that is not critical to the image. You can also pass a comma-separated list of specific metadata chunks to remove. -s can be used as a shorthand for --strip safe.
• Alpha: --alpha can improve compression of images with transparency, by altering the color values of fully transparent pixels. This is generally recommended, but take care as this is technically a lossy transformation and may be unsuitable for some specific applications.

More advanced options can be found by running oxipng --help, or viewed here.

Some options have both short (-a) and long (--alpha) forms. Which form you use is just a matter of preference. Multiple short options can be combined together, e.g.: -savvo6 is equivalent to to --strip safe --alpha --verbose --verbose --opt 6. Note that all options are case-sensitive.

Create a .pre-commit-config.yaml file like this, or add the lines after the repos map preamble to an already existing one:

repos:
  - repo: https://github.com/shssoichiro/oxipng
    rev: v9.0.0
    hooks:
      - id: oxipng
        args: ["-o", "4", "--strip", "safe", "--alpha"]

Trunk is an extendable superlinter which can be used to run oxipng to automatically optimize pngs when committing them into a git repo, or to gate any pngs being added to a git repo on whether they are optimized. The trunk oxipng integration is here.

To enable oxipng via trunk:

# to get the latest version:
trunk check enable oxipng

# to get a specific version:
trunk check enable [email protected]

or modify .trunk/trunk.yaml in your repo to contain:

lint:
  enabled:
    - [email protected]

Then just run:

# to optimize a png:
trunk fmt <file>

# to check if a png is already optimized:
trunk check <file>

You can setup trunk to manage your git hooks and automatically optimize any pngs you commit to git, when you git commit. To enable this, run:

trunk actions enable trunk-fmt-pre-commit

Although originally intended to be used as an executable, oxipng can also be used as a library in other Rust projects. To do so, simply add oxipng as a dependency in your Cargo.toml, then extern crate oxipng in your project. You should then have access to all of the library functions documented here. The simplest method of usage involves creating an Options struct and passing it, along with an input filename, into the optimize function.

It is recommended to disable the "binary" feature when including oxipng as a library. Currently, there is no simple way to just disable one feature in Cargo, it has to be done by disabling default features and specifying the desired ones, for example: oxipng = { version = "9.0", features = ["parallel", "zopfli", "filetime"], default-features = false }

Oxipng began as a complete rewrite of the OptiPNG project, which was assumed to be dead as no commit had been made to it since March 2014. (OptiPNG has since released a new version, after Oxipng was first released.) The name has been changed to avoid confusion and potential legal issues.

The core goal of rewriting OptiPNG was to implement multithreading, which would be very difficult to do within the existing C codebase of OptiPNG. This also served as an opportunity to choose a more modern, safer language (Rust).

Note that, while similar, Oxipng is not a drop-in replacement for OptiPNG. If you are migrating from OptiPNG, please check the help before using.

Any contributions are welcome and will be accepted via pull request on GitHub. Bug reports can be filed via GitHub issues. Please include as many details as possible. If you have the capability to submit a fix with the bug report, it is preferred that you do so via pull request, however you do not need to be a Rust developer to contribute. Other contributions (such as improving documentation or translations) are also welcome via GitHub.

Oxipng is open-source software, distributed under the MIT license.

Tested OxiPNG 9.0.0 (commit c16519b38b0519988db625913be919d4f0e42f5d, compiled on rustc 1.74.0-nightly (7b4d9e155 2023-09-28)) against OptiPNG version 0.7.7, as packaged by Debian unstable, on a Linux 6.5.0-2-amd64 kernel, Intel Core i7-12700 CPU (8 performance cores, 4 efficiency cores, 20 threads), DDR5-5200 RAM in dual channel configuration.

Benchmark 1: ./target/release/oxipng -P ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):      59.6 ms ±   7.7 ms    [User: 77.4 ms, System: 3.6 ms]
  Range (min … max):    53.3 ms …  89.9 ms    32 runs

Benchmark 2: optipng -simulate ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):     132.4 ms ±   0.8 ms    [User: 132.5 ms, System: 0.6 ms]
  Range (min … max):   131.8 ms … 134.4 ms    22 runs

Summary
  ./target/release/oxipng -P ./tests/files/rgb_16_should_be_grayscale_8.png ran
    2.22 ± 0.29 times faster than optipng -simulate ./tests/files/rgb_16_should_be_grayscale_8.png

Benchmark 1: ./target/release/oxipng -o4 -P ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):      88.7 ms ±   4.3 ms    [User: 270.3 ms, System: 11.0 ms]
  Range (min … max):    86.8 ms … 109.4 ms    26 runs

Benchmark 2: optipng -o 4 -simulate ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):     444.9 ms ±   0.3 ms    [User: 444.8 ms, System: 0.7 ms]
  Range (min … max):   444.4 ms … 445.6 ms    10 runs

Summary
  ./target/release/oxipng -o4 -P ./tests/files/rgb_16_should_be_grayscale_8.png ran
    5.01 ± 0.25 times faster than optipng -o 4 -simulate ./tests/files/rgb_16_should_be_grayscale_8.png

Benchmark #1: ./target/release/oxipng -P ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):     128.8 ms ±  14.2 ms    [User: 296.0 ms, System: 14.3 ms]
  Range (min … max):    98.8 ms … 152.3 ms    21 runs

Benchmark #2: optipng -simulate ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):     254.2 ms ±  16.0 ms    [User: 252.8 ms, System: 1.2 ms]
  Range (min … max):   208.4 ms … 263.8 ms    14 runs

Summary
  './target/release/oxipng -P ./tests/files/rgb_16_should_be_grayscale_8.png' ran
    1.97 ± 0.25 times faster than 'optipng -simulate ./tests/files/rgb_16_should_be_grayscale_8.png'



Benchmark #1: ./target/release/oxipng -o4 -P ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):     141.4 ms ±  14.9 ms    [User: 611.7 ms, System: 21.1 ms]
  Range (min … max):   100.2 ms … 160.4 ms    23 runs

Benchmark #2: optipng -o 4 -simulate ./tests/files/rgb_16_should_be_grayscale_8.png
  Time (mean ± σ):     730.0 ms ±  25.9 ms    [User: 728.0 ms, System: 1.2 ms]
  Range (min … max):   713.3 ms … 768.2 ms    10 runs

Summary
  './target/release/oxipng -o4 -P ./tests/files/rgb_16_should_be_grayscale_8.png' ran
    5.16 ± 0.58 times faster than 'optipng -o 4 -simulate ./tests/files/rgb_16_should_be_grayscale_8.png'