OpenCV python wheels built against CUDA 12.5, Nvidia Video Codec SDK 12.2 and cuDNN 9.2.0.
Suitable for all devices of compute capability >= 5.0 with binary compatible code for devices of compute capability 5.0-9.0.
Nvidia GPU Computing Toolkit v12.5 is required for import cv2 to work and cuDNN 9.2.0 for accelerated inference when using the dnn module.
Note Windows OS: This wheel relies on cuDNN being installed in the CUDA Toolkit directory. Therefore you can either
-
Download the cuDNN Tarball (not the installer) and extract its contents to your CUDA directory, or
-
Add the path to the bin folder inside the cuDNN installation directory to your PATH_TO_PYTHON_DIST/Lib/site-packages/cv2/config.py file. e.g.
import os
BINARIES_PATHS = [
os.path.join('D:/build/opencv/install', 'x64/vc17/bin'),
os.path.join(os.getenv('CUDA_PATH', 'C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v12.5'), 'bin')
os.path.join('C:/Program Files/NVIDIA/CUDNN/v9.2/bin/12.5')
] + BINARIES_PATHS
Build Summary
Windows
Nvidia CMake configuration output
NVIDIA CUDA: YES (ver 12.5, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 60 61 70 75 80 86 89 90
NVIDIA PTX archs: 90cuDNN: YES (ver 9.2.0)
Build commands
set "CMAKE_ARGS=-DWITH_CUDA=ON -DCUDA_ARCH_BIN=5.0;5.2;6.0;6.1;7.0;7.5;8.0;8.6;8.9;9.0 -DCUDA_ARCH_PTX=9.0"
set ENABLE_CONTRIB=1
python.exe setup.py bdist_wheel --py-limited-api=cp37
Ubuntu 22.04
Nvidia CMake configuration output
NVIDIA CUDA: YES (ver 12.5, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 53 60 61 62 70 72 75 80 86 87 89 90
NVIDIA PTX archs: 90cuDNN: YES (ver 9.2.0)
Build commands
export CMAKE_ARGS="-DWITH_CUDA=ON -DCUDA_ARCH_BIN=5.0;5.2;5.3;6.0;6.1;6.2;7.0;7.2;7.5;8.0;8.6;8.7;8.9;9.0 -DCUDA_ARCH_PTX=9.0"
export ENABLE_CONTRIB=1
python setup.py bdist_wheel --py-limited-api=cp37