pngquant - Man Page
PNG converter and lossy image compressor
Examples (TL;DR)
Compress a specific PNG as much as possible and write result to a new file:
pngquant path/to/file.png
Compress a specific PNG and override original:
pngquant --ext .png --force path/to/file.png
Try to compress a specific PNG with custom quality (skip if below the min value):
pngquant --quality 0-100 path/to/file.png
Compress a specific PNG with the number of colors reduced to 64:
pngquant 64 path/to/file.png
Compress a specific PNG and skip if the file is larger than the original:
pngquant --skip-if-larger path/to/file.png
Compress a specific PNG and remove metadata:
pngquant --strip path/to/file.png
Compress a specific PNG and save it to the given path:
pngquant path/to/file.png --output path/to/file.png
Compress a specific PNG and show progress:
pngquant --verbose path/to/file.png
Synopsis
pngquant | ⟨options⟩ [ncolors] file [file ...] |
pngquant | ⟨options⟩ [ncolors] - < file > file |
Description
pngquant converts 32-bit RGBA PNGs to 8-bit (or smaller) RGBA-palette PNGs, optionally using Floyd-Steinberg dithering. The output filename is the same as the input name except that it ends in ‘-fs8.png
’ or ‘-or8.png
’ (unless the input is stdin, in which case the quantized image will go to stdout). The default behavior if the output file exists is to skip the conversion; use --force to overwrite.
Options
- -o out.png, --output out.png
Writes converted file to the given path. When this option is used only single input file is allowed.
- --ext new.png
File extension (suffix) to use for output files instead of the default ‘
-fs8.png
’ or ‘-or8.png
’.- -f, --force
Overwrite existing output files. “--ext .png --force” can be used to convert files in place (which is unsafe).
- --nofs, --ordered
Disable Floyd-Steinberg dithering.
- --floyd [=N]
Set dithering level using fractional number between
0
(none) and1
(full, the default).- -s N, --speed N
1
(brute-force) to11
(fastest). The default is3
. Speed10
has 5% lower quality, but is about 8 times faster than the default. Speed 11 disables dithering and lowers compression level.- -Q min-max, --quality min-max
min and max are numbers in range
0
(worst) to100
(perfect), similar to JPEG. pngquant will use the least amount of colors required to meet or exceed the max quality. If conversion results in quality below the min quality the image won't be saved (or if outputting to stdin, 24-bit original will be output) and pngquant will exit with status code99
.- --skip-if-larger
If conversion results in a file larger than the original, the image won't be saved and pngquant will exit with status code
98
. Additionally, file size gain must be greater than the amount of quality lost. If quality drops by 50%, it will expect 50% file size reduction to consider it worthwhile.- --posterize bits
Truncate number of least significant bits of color (per channel). Use this when image will be output on low-depth displays (e.g. 16-bit RGB). pngquant will make almost-opaque pixels fully opaque and will reduce amount of semi-transparent colors. When this option is enabled the default filename suffix is ‘
-ie-fs8.png
’ / ‘-ie-or8.png
’.- --strip
Remove optional chunks (metadata) from PNG files.
- --transbug
Workaround for readers that expect fully transparent color to be the last entry in the palette.
- -v, --verbose
Enable verbose messages showing progress and information about input/output. Opposite is
--quiet
. Errors are output tostderr
regardless of this option.- -V, --version
Display version on
stdout
and exit.- -h, --help
Display help and exit.
Example
Creating a new image with the number of colors reduced to 64:
pngquant
64 image.png
The resulting image will have 64 colors and will be saved as image-fs8.png
.
Overwriting image in-place if it can be reduced without too much quality loss:
Author
pngquant is developed by Kornel Lesinski <kornel@pngquant.org> based on code by Greg Roelofs <newt@pobox.com>.