What Operating System(s) are you seeing this problem on?
Windows
dlib version
19.24
Python version
3.13
Compiler
MSVC toolchain via pip (Windows build environment, triggered from VS Code terminal)
Expected Behavior
dlib should install successfully using pip without requiring manual compilation or getting stuck while building wheel.
Current Behavior
When running:
pip install opencv-python face_recognition numpy
the installation gets stuck at:
Building wheel for dlib (pyproject.toml) ...
The process either takes extremely long or does not complete. In some cases, it is manually cancelled.
This blocks the installation of face_recognition because it depends on dlib.
Steps to Reproduce
- Open VS Code terminal on Windows
- Run: pip install opencv-python face_recognition numpy
- pip starts resolving dependencies
- It begins building dlib from source
- Installation gets stuck at "Building wheel for dlib"
Anything else?
Environment:
- Windows OS
- Python 3.13
- pip install executed inside VS Code terminal
It seems dlib does not have prebuilt wheels for Python 3.13 yet, causing compilation from source which leads to long or failed builds.
This also affects installation of face_recognition since it depends on dlib.
A workaround or prebuilt binary would be helpful.
What Operating System(s) are you seeing this problem on?
Windows
dlib version
19.24
Python version
3.13
Compiler
MSVC toolchain via pip (Windows build environment, triggered from VS Code terminal)
Expected Behavior
dlib should install successfully using pip without requiring manual compilation or getting stuck while building wheel.
Current Behavior
When running:
pip install opencv-python face_recognition numpythe installation gets stuck at:
Building wheel for dlib (pyproject.toml) ...
The process either takes extremely long or does not complete. In some cases, it is manually cancelled.
This blocks the installation of face_recognition because it depends on dlib.
Steps to Reproduce
Anything else?
Environment:
It seems dlib does not have prebuilt wheels for Python 3.13 yet, causing compilation from source which leads to long or failed builds.
This also affects installation of face_recognition since it depends on dlib.
A workaround or prebuilt binary would be helpful.