Skip to content

Releases: cudawarped/opencv-python-cuda-wheels

4.8.0.20230804

04 Aug 13:46
43cd716
Compare
Choose a tag to compare

OpenCV python wheels built against CUDA 12.2, Nvidia Video Codec SDK 12.1 and cuDNN 8.9.3.

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 v12.2 is required for import cv2 to work and cuDNN 8.9.3 for accelerated inference when using the dnn module.

Build Summary

Windows

Nvidia CMake configuration output

NVIDIA CUDA: YES (ver 12.2, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 60 61 70 75 80 86 89 90
NVIDIA PTX archs: 90

cuDNN: YES (ver 8.9.3)

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
set ENABLE_ROLLING=1
python.exe setup.py bdist_wheel --py-limited-api=cp36

Linux

Nvidia CMake configuration output

NVIDIA CUDA: YES (ver 12.2, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 53 60 61 62 70 72 75 80 86 87 89 90
NVIDIA PTX archs: 90

cuDNN: YES (ver 8.9.3)

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
export ENABLE_ROLLING=1
python setup.py bdist_wheel --py-limited-api=cp36

4.8.0.74

02 Jul 15:47
0ec125d
Compare
Choose a tag to compare

OpenCV python wheels built against CUDA 12.1 Nvidia Video Codec SDK 12.1 and cuDNN 8.9.2.

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 v12.1 is required for import cv2 to work and cuDNN 8.9.2 for accelerated inference when using the dnn module.

Build Summary

Windows

Nvidia CMake configuration output

NVIDIA CUDA: YES (ver 12.1, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 60 61 70 75 80 86 89 90
NVIDIA PTX archs: 90

cuDNN: YES (ver 8.9.2)

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=cp36

Linux

Nvidia CMake configuration output

NVIDIA CUDA: YES (ver 12.1, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 53 60 61 62 70 72 75 80 86 87 89 90
NVIDIA PTX archs: 90

cuDNN: YES (ver 8.9.2)

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=cp36

4.7.0.20230527

27 May 07:28
6b73d90
Compare
Choose a tag to compare

OpenCV python wheels built against CUDA 12.1 Nvidia Video Codec SDK 12.1 and cuDNN 8.9.1.

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 v12.1 is required for import cv2 to work and cuDNN 8.9.1 for accelerated inference when using the dnn module.

Build Summary

Windows

Nvidia CMake configuration output

NVIDIA CUDA: YES (ver 12.1, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 60 61 70 75 80 86 89 90
NVIDIA PTX archs: 90

cuDNN: YES (ver 8.9.1)

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
set ENABLE_ROLLING=1
python.exe setup.py bdist_wheel --py-limited-api=cp36

Linux

Nvidia CMake configuration output

NVIDIA CUDA: YES (ver 12.1, CUFFT CUBLAS NVCUVID NVCUVENC)
NVIDIA GPU arch: 50 52 53 60 61 62 70 72 75 80 86 87 89 90
NVIDIA PTX archs: 90

cuDNN: YES (ver 8.9.1)

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
export ENABLE_ROLLING=1
python setup.py bdist_wheel --py-limited-api=cp36

4.7.0.20221229

29 Dec 20:27
7ebbdc3
Compare
Choose a tag to compare

OpenCV python wheels built against CUDA 12.0 Nvidia Video Codec SDK 12.0 and cuDNN 8.8.1.

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 v12.0 is required for import cv2 to work and cuDNN 8.7.0 for accelerated inference when using the dnn module.

CMAKE_ARGS

Windows

-DWITH_CUDA=ON
-DCUDA_FAST_MATH=ON
-DWITH_CUFFT=ON
-DWITH_CUBLAS=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
-DWITH_NVCUVID=ON
-DWITH_NVCUVENC=ON

set "CMAKE_ARGS=-DWITH_CUDA=ON -DCUDA_FAST_MATH=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
set ENABLE_ROLLING=1
python.exe setup.py bdist_wheel --py-limited-api=cp36

Linux

-DWITH_CUDA=ON
-DCUDA_FAST_MATH=ON
-DWITH_CUFFT=ON
-DWITH_CUBLAS=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
-DWITH_NVCUVID=ON
-DWITH_NVCUVENC=ON

export CMAKE_ARGS="-DWITH_CUDA=ON -DCUDA_FAST_MATH=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
export ENABLE_ROLLING=1
python setup.py bdist_wheel --py-limited-api=cp36

4.6.0.20221107

07 Nov 12:33
3725898
Compare
Choose a tag to compare

Winows OpenCV python wheel built against CUDA 11.8 Nvidia Video Codec SDK 11.1 and cuDNN 8.6.0.

Suitable for all devices of compute capability >= 3.5 with binary compatible code for devices of compute capability 3.5-8.6.

Nvidia GPU Computing Toolkit v11.8 is required for import cv2 to work and cuDNN 8.6.0 for accelerated inference when using the dnn module.

CMAKE_ARGS

-DWITH_CUDA=ON
-DCUDA_FAST_MATH=ON
-DWITH_CUBLAS=ON
-DCUDA_ARCH_PTX=8.6
-DWITH_NVCUVID=ON
-DWITH_NVCUVENC=ON

Windows

set CMAKE_ARGS="-DWITH_CUDA=ON -DCUDA_FAST_MATH=ON -DWITH_CUBLAS=ON -DCUDA_ARCH_PTX=8.6 -DWITH_NVCUVID=ON -DWITH_NVCUVENC=ON"
set ENABLE_CONTRIB=1
set ENABLE_ROLLING=1
python.exe setup.py bdist_wheel --py-limited-api=cp36

4.6.0.20221102

03 Nov 09:00
3725898
Compare
Choose a tag to compare

Linux OpenCV python wheel built against CUDA 11.7, Nvidia Video Codec SDK 11.1 and cuDNN 8.4.1.

Suitable for all devices of compute capability >= 3.5 with binary compatible code for devices of compute capability 3.5-8.6.

Nvidia GPU Computing Toolkit v11.7 is required for import cv2 to work and cuDNN 8.4.1 for accelerated inference when using the dnn module.

CMAKE_ARGS

-DWITH_CUDA=ON
-DCUDA_FAST_MATH=ON
-DWITH_CUBLAS=ON
-DCUDA_ARCH_PTX=8.6
-DWITH_NVCUVID=ON
-DWITH_NVCUVENC=ON

Linux

export CMAKE_ARGS="-DWITH_CUDA=ON -DCUDA_FAST_MATH=ON -DWITH_CUBLAS=ON -DCUDA_ARCH_PTX=8.6 -DWITH_NVCUVID=ON -DWITH_NVCUVENC=ON"
export ENABLE_CONTRIB=1
export ENABLE_ROLLING=1
python setup.py bdist_wheel --py-limited-api=cp36

4.6.0.20220923

23 Sep 09:24
3725898
Compare
Choose a tag to compare

Fork with python bindings for cuda::nonLocalMeans, cuda::fastNlMeansDenoising, cuda::fastNlMeansDenoisingColored.

OpenCV python wheel built against CUDA 11.7, Nvidia Video Codec SDK 11.1 and cuDNN 8.4.1.

Suitable for all devices of compute capability >= 3.5 with binary compatible code for devices of compute capability 3.5-8.6.

Nvidia GPU Computing Toolkit v11.7 is required for import cv2 to work and cuDNN 8.4.1 for accelerated inference when using the dnn module.
CMAKE_ARGS

4.6.0.20220920

20 Sep 10:56
3725898
Compare
Choose a tag to compare

OpenCV python wheel built against CUDA 11.7, Nvidia Video Codec SDK 11.1 and cuDNN 8.4.1.

Suitable for all devices of compute capability >= 3.5 with binary compatible code for devices of compute capability 3.5-8.6.

Nvidia GPU Computing Toolkit v11.7 is required for import cv2 to work and cuDNN 8.4.1 for accelerated inference when using the dnn module.

CMAKE_ARGS

-DWITH_CUDA=ON
-DCUDA_FAST_MATH=ON
-DWITH_CUBLAS=ON
-DCUDA_ARCH_PTX=8.6
-DWITH_NVCUVID=ON
-DBUILD_opencv_world=ON

Windows

set CMAKE_ARGS="-DBUILD_opencv_world=ON -DWITH_CUDA=ON -DCUDA_FAST_MATH=ON -DWITH_CUBLAS=ON -DCUDA_ARCH_PTX=8.6 -DWITH_NVCUVID=ON"

Linux

export CMAKE_ARGS="-DBUILD_opencv_world=ON -DWITH_CUDA=ON -DCUDA_FAST_MATH=ON -DWITH_CUBLAS=ON -DCUDA_ARCH_PTX=8.6 -DWITH_NVCUVID=ON"

OpenCV 4.6.0+

15 Sep 15:24
3725898
Compare
Choose a tag to compare

OpenCV python wheel built against CUDA 11.7, Nvidia Video Codec SDK 11.1 and cuDNN 8.4.1.

Suitable for all devices of compute capability >= 3.5 with binary compatible code for devices of compute capability 3.5-8.6.

Nvidia GPU Computing Toolkit v11.7 is required for import cv2 to work and cuDNN 8.4.1 for accelerated inference when using the dnn module.

CMAKE_ARGS

-DWITH_CUDA=ON
-DCUDA_FAST_MATH=ON
-DWITH_CUBLAS=ON
-DCUDA_ARCH_PTX=8.6
-DWITH_NVCUVID=ON

Windows

set CMAKE_ARGS="-DWITH_CUDA=ON -DCUDA_FAST_MATH=ON -DWITH_CUBLAS=ON -DCUDA_ARCH_PTX=8.6 -DWITH_NVCUVID=ON"