Jpegli

Spec sheet

Purpose

Jpegli is an improved JPEG encoder and decoder implementation with API and ABI compatibility with libjpeg62.Read 12 Oct 2026 · github.com

Intended users

The README identifies the Git guide as being for developers and provides build and testing instructions for the project.Read 11 Oct 2026 · github.com

Free plan

YesRead 24 Sep 2026 · github.com

Support

The repository asks users to report bugs or other issues by opening a GitHub issue.Read 12 Oct 2026 · github.com

License

The repository lists the BSD-3-Clause license.Read 12 Oct 2026 · github.com

Founded

2024Read 19 Oct 2026 · github.com

Adaptive quantization

Jpegli uses adaptive dead-zone quantization with higher zero-coefficient thresholds in noisy image regions than in smoother regions.Read 12 Oct 2026 · github.com

API access

YesRead 24 Sep 2026 · github.com

Build outputs

Building the project produces cjpegli and djpegli binaries and a libjpeg.so.62.3.0 shared library that can replace a system library with the same name.Read 11 Oct 2026 · github.com

Build targets

The build documentation describes cross-compiling for other architectures and mentions additional environment setup for Windows targets.Read 11 Oct 2026 · github.com

Command line and library

Building the project produces cjpegli and djpegli binaries and a libjpeg shared library intended as a drop-in replacement for the system library of the same name.Read 12 Oct 2026 · github.com

Compatibility

Google describes Jpegli as a fully interoperable encoder and decoder that complies with the original JPEG standard and its conventional 8-bit format.Read 12 Oct 2026 · opensource.googleblog.com

Compression

Google says Jpegli can compress high quality images 35% more than traditional JPEG codecs.Read 12 Oct 2026 · opensource.googleblog.com

Compression modes

lossyRead 24 Sep 2026 · github.com

Decoder output

The decoder supports 16-bit unsigned and 32-bit floating-point output buffers.Read 11 Oct 2026 · github.com

Decoder outputs

The decoder supports 16-bit unsigned and 32-bit floating point output buffers.Read 12 Oct 2026 · github.com

Encoder input

The encoder supports 16-bit unsigned and 32-bit floating-point input buffers.Read 11 Oct 2026 · github.com

Encoder inputs

The encoder supports 16-bit unsigned and 32-bit floating point input buffers.Read 12 Oct 2026 · github.com

Floating-point processing

Color-space conversion, chroma subsampling, and DCT use floating-point precision before final quantization.Read 11 Oct 2026 · github.com

Integration

The library is designed as a drop-in replacement compatible with the libjpeg62 API and ABI.Read 11 Oct 2026 · github.com

macOS build

The repository includes separate instructions for building on macOS.Read 12 Oct 2026 · github.com

Patent grant

Google grants a perpetual, worldwide, non-exclusive, no-charge, royalty-free patent license for specified claims necessarily infringed by the implementation, subject to stated termination conditions.Read 11 Oct 2026 · github.com

Quality control

Encoding quality can be set with a distance parameter analogous to JPEG XL’s distance parameter.Read 12 Oct 2026 · github.com

Supported formats

JPEGRead 24 Sep 2026 · github.com

Windows build

The project’s Windows instructions describe building Jpegli in MSYS2 on Windows.Read 12 Oct 2026 · github.com

XYB color

Jpegli can more efficiently compress JPEGs with an ICC profile representing XYB by using specialized quantization tables for the original X, Y, and B channels.Read 11 Oct 2026 · github.com

XYB colorspace

For JPEGs with an ICC profile representing the XYB colorspace, Jpegli can use specialized quantization tables without converting the image to YCbCr.Read 12 Oct 2026 · github.com

Plans and prices

Every Jpegli plan and price →

Where Specifiction places it

• Image Compression Software#29 of 858.3