Download AnyLabeling

Get the current stable release, v0.4.43, for your platform. Choose CPU if you are unsure; choose Accelerated when you have the matching hardware.

Windows x64

Windows

NVIDIA CUDA acceleration is included in the GPU build.

Download CPU Download GPU

Linux x64

Linux

NVIDIA CUDA acceleration is included in the GPU build.

Download CPU Download GPU

Apple Silicon

macOS

CoreML acceleration is included in the GPU archive.

Download CPU Download CoreML

Validated release. All six v0.4.43 artifacts publish SHA-256 digests and were launch-smoke-tested on native Linux, Windows, and macOS runners before upload. Avoid the superseded v0.4.40 macOS and Linux artifacts.

Install with Python

Python 3.11 or newer is required; Python 3.12 is recommended. Use a dedicated environment.

conda create -n anylabeling python=3.12
conda activate anylabeling
python -m pip install anylabeling
anylabeling

For NVIDIA CUDA on Linux or Windows, install anylabeling-gpu in a fresh environment. Apple Silicon users can install anylabeling[macos] for CoreML. See Installation and Hardware Acceleration for complete setup instructions.

Need source archives, checksums, or an older version? View all GitHub releases →