cuda-bindings
13.0.0 Release notes¶
Released on Aug 6, 2025
Highlights¶
Support CUDA 13.0.
A utility module
cuda.bindings.utils
is addedUsing
int(cuda_obj)
to retrieve the underlying address of a CUDA object is deprecated and subject to future removal. Please switch to useget_cuda_native_handle()
instead.
The
cuda.bindings.cufile
Python module was added, wrapping the cuFile C APIs. Supported on Linux only.Currently using this module requires NumPy to be present. Any recent NumPy 1.x or 2.x should work.
Python bindings in every module, including
driver
,runtime
, andnvrtc
, now have the GIL released before calling the underlying C APIs.
Breaking changes¶
For breaking changes in the CUDA APIs, please see the CUDA 13.0 release notes.
The trampoline modules
cuda.{cuda,cudart,nvrtc}
are now removed. Users should switch to usecuda.bindings.{driver,runtime,nvrtc}
instead.
Bug fixes¶
Fix a library loading bug that preferred shared libraries without a SOVERSION.
Miscellaneous¶
All Python bindings now have the GIL released when calling into the underlying C APIs.
Added PTX utilities including
get_minimal_required_cuda_ver_from_ptx_ver()
andget_ptx_ver()
.Common CUDA objects such as
cudaStream_t
now compare equal if the underlying address is the same.Add a binding to
nvvmGetErrorString()
.Build the bindings with Cython profile hooks disabled.
The internal pathfinder module is now isolated to a standalone package
cuda-pathfinder
and made as a required dependency.
Known issues¶
Updating from older versions (v12.6.2.post1 and below) via
pip install -U cuda-python
might not work. Please do a clean re-installation by uninstallingpip uninstall -y cuda-python
followed by installingpip install cuda-python
.