Skip to content

AnyLabeling v0.4.42

Choose a tag to compare

View all tags
@github-actions github-actions released this 29 Aug 18:56
· 10 commits to main since this release
v0.4.42
8fe3c5b
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Highlights

v0.4.42 is the stable replacement for the affected v0.4.40 artifacts. It combines cross-platform accelerator support with corrected macOS/Linux packaging and release-time launch gates.

Fixed

  • Re-enabled the model picker after an invalid custom model fails to load.
  • Preserved required PyInstaller symlinks in macOS ZIP archives.
  • Bundled the matching libexpat in Linux frozen applications.

Hardware acceleration

  • NVIDIA CUDA on Linux and Windows; the GPU distribution includes pip-managed CUDA 12 and cuDNN runtime libraries.
  • CoreML on Apple Silicon.
  • Opt-in DirectML, OpenVINO, TensorRT, ROCm, MIGraphX, WebGPU, and vendor NPU providers through ANYLABELING_DEVICE.
  • CPU fallback when a requested provider is unavailable.

Downloads

Choose the artifact for your platform:

  • Linux: AnyLabeling-Linux-CPU-x64 or AnyLabeling-Linux-GPU-x64
  • Windows: AnyLabeling-Windows-CPU-x64.exe or AnyLabeling-Windows-GPU-x64.exe
  • Apple Silicon macOS: AnyLabeling-macOS-CPU.zip or AnyLabeling-macOS-GPU.zip

Python packages:

python -m pip install anylabeling==0.4.42
# NVIDIA CUDA on Linux/Windows, in a separate environment:
python -m pip install anylabeling-gpu==0.4.42

Validation

  • All six uploaded artifacts were checksum-verified against the downloaded release assets.
  • CPU and accelerated applications were launch-tested on Linux, Windows, and Apple Silicon macOS.
  • CUDA, DirectML, CoreML, OpenVINO, and CPU provider paths were exercised in dedicated environments on available hardware.
  • Release workflows now launch frozen Linux/Windows applications and extracted macOS archives before upload.

Full changelog: v0.4.40...v0.4.42

Assets 8
Loading