cuda-bindings
13.X.Y Release notes#
Released on TBD
Highlights#
Migrated wheel dependencies from individual NVIDIA packages to the
cuda-toolkit
metapackage for improved dependency resolution and version constraints.Automatic CUDA library path detection based on
CUDA_HOME
, eliminating the need to manually setLIBRARY_PATH
environment variables for installation.The
[all]
optional dependencies now usecuda-toolkit
with appropriate extras instead of individual packages. The NVCC compiler is no longer automatically installed withpip install cuda-python[all]
as it was previously included only to access the NVVM library, which now has its own dedicated wheel. Users who need the NVCC compiler should explicitly install it withpip install cuda-toolkit[nvcc]==X.Y
with the appropriate version for their needs.The Python overhead of calling functions in CUDA bindings in
driver
,runtime
andnvrtc
has been reduced by approximately 30%.On Windows, the
pywin32
dependency has been removed. The necessary Windows API functions are now accessed directly.Updated the
cuda.bindings.runtime
module to statically link against the CUDA Runtime library from CUDA Toolkit 13.0.1.cyruntime.getLocalRuntimeVersion
now uses pathfinder to find the CUDA runtime.
Bug fixes#
Restoring the
cuCheckpointProcessRestore()
API removed by mistake.
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
.