From @ralfgommers on the #NumPy issue tracker regarding #NumPy2:

Public service announcement: if your package uses pybind11 to interact with NumPy, you have to upgrade to the just-released pybind11 2.12.0: https://pypi.org/project/pybind11/2.12.0/.

The numpy 2.0.0rc1 is imminent, and it will not work with older pybind11 versions.

https://github.com/numpy/numpy/issues/24300#issuecomment-2027392563

pybind11

Seamless operability between C++11 and Python

PyPI

We recently prepared #scikitimage for #numpy2, which introduces a few significant changes to the promotion rules for scalars and arrays #nep50. So I thought to write up a few takeaways that hopefully make life a bit easier for you and your #numpy code:

https://grueter.dev/blog/preparing-for-numpy-2/

Tips on preparing your Python code for NumPy 2 · Lars Grüter

Lars Grüter. Personal website and blog.

#NumPy2 is coming! Ensure your code doesn't break by avoiding NumPy 2 installation, pinning dependencies, monitoring updates, and using Ruff to update code. #Python #AISummary
https://pythonspeed.com/articles/numpy-2/
NumPy 2 is coming: preventing breakage, updating your code

NumPy 2 is coming, and it’s backwards incompatible. Learn how to keep your code from breaking, and how to upgrade.

Python⇒Speed