nvidia/cuda Docker Image Overview

nvidia/cuda

英伟达在GitLab仓库（gitlab.com/nvidia/cuda）提供的CUDA（并行计算平台）和cuDNN（深度神经网络加速库）镜像，为开发者提供了预配置的开发环境，支持高效进行并行计算应用开发、深度学习模型训练及推理任务，确保了环境的一致性和部署的便捷性，是构建基于英伟达GPU加速应用的重要资源。

1946 收藏0 次下载activenvidia镜像

🚀专业版镜像服务，面向生产环境设计

版本下载

🚀专业版镜像服务，面向生产环境设计

NVIDIA CUDA

CUDA is a parallel computing platform and programming model developed by NVIDIA for general computing on graphical processing units (GPUs). With CUDA, developers can dramatically speed up computing applications by harnessing the power of GPUs.

The CUDA Toolkit from NVIDIA provides everything you need to develop GPU-accelerated applications. The CUDA Toolkit includes GPU-accelerated libraries, a compiler, development tools and the CUDA runtime.

The CUDA container images provide an easy-to-use distribution for CUDA supported platforms and architectures.

End User License Agreements

The images are governed by the following NVIDIA End User License Agreements. By pulling and using the CUDA images, you accept the terms and conditions of these licenses. Since the images may include components licensed under open-source licenses such as GPL, the sources for these components are archived here.

NVIDIA Deep learning Container License

To view the NVIDIA Deep Learning Container license, click here

Documentation

For more information on CUDA, including the release notes, programming model, APIs and developer tools, visit the CUDA documentation site.

Announcement

CUDA Container Support Policy

CUDA image container tags have a lifetime. The tags will be deleted Six Months after the last supported "Tesla Recommended Driver" has gone end-of-life OR a newer update release has been made for the same CUDA version.

Please see CUDA Container Support Policy for more information.

Breaking changes are announced on Gitlab Issue #209.

Cuda Repo Signing Key has Changed!

This may present itself as the following errors.

debian:

Reading package lists... Done
W: GPG error: [***]  InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY A4B469963BF863CC
W: The repository '[***]  InRelease' is not signed.
N: Data from such a repository can't be authenticated and is therefore potentially dangerous to use.
N: See apt-secure(8) manpage for repository creation and user configuration details.

RPM:

warning: /var/cache/dnf/cuda-fedora32-x86_64-d60aafcddb176bf5/packages/libnvjpeg-11-1-11.3.0.105-1.x86_64.rpm: Header V4 RSA/SHA512 Signature, key ID d42d0685: NOKEY
cuda-fedora32-x86_64                                                                                  23 kB/s | 1.6 kB     00:00
Importing GPG key 0x7FA2AF80:
 Userid     : "cudatools <***>"
 Fingerprint: AE09 FE4B BD22 3A84 B2CC FCE3 F60F 4B3D 7FA2 AF80
 From       : [***]
Is this ok [y/N]: y
Key imported successfully
Import of key(s) didn't help, wrong key(s)?
Public key for libnvjpeg-11-1-11.3.0.105-1.x86_64.rpm is not installed. Failing package is: libnvjpeg-11-1-11.3.0.105-1.x86_64
 GPG Keys are configured as: [***]
The downloaded packages were saved in cache until the next successful transaction.
You can remove cached packages by executing 'dnf clean packages'.
Error: GPG check FAILED

Updated images will be pushed out over the next few days containing the new repo key. Please follow progress using the links below:

[***]
[***]

Multi-arch image manifests are now LIVE for all supported CUDA container image versions

It is now possible to build CUDA container images for all supported architectures using Docker Buildkit in one step. See the example script below.

The deprecated image names nvidia/cuda-arm64 and nvidia/cuda-ppc64le will remain available, but no longer supported.

The following product pages still exist but will no longer be supported:

[***]
[***]

The following gitlab repositories will be archived:

[***]

Deprecated: "latest" tag

The "latest" tag for CUDA, CUDAGL, and OPENGL images has been deprecated on NGC and Docker Hub.

With the removal of the latest tag, the following use case will result in the "manifest unknown" error:

$ docker pull nvidia/cuda
Error response from daemon: manifest for nvidia/cuda:latest not found: manifest unknown: manifest
unknown

This is not a bug.

Overview of Images

Three flavors of images are provided:

base: Includes the CUDA runtime (cudart)
runtime: Builds on the base and includes the CUDA math libraries, and NCCL. A runtime image that also includes cuDNN is available. Some images may also include TensorRT.
devel: Builds on the runtime and includes headers, development tools for building CUDA images. These images are particularly useful for multi-stage builds.

The Dockerfiles for the images are open-source and licensed under 3-clause BSD. For more information see the Supported Tags section below.