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。
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: http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY A4B469963BF863CC W: The repository 'http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 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 <cudatools@nvidia.com>" Fingerprint: AE09 FE4B BD22 3A84 B2CC FCE3 F60F 4B3D 7FA2 AF80 From : https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/7fa2af80.pub 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: https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/7fa2af80.pub 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:
- https://hub.docker.com/r/nvidia/cuda-ppc64le
- https://hub.docker.com/r/nvidia/cuda-arm64
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 thebaseand includes the CUDA math libraries, and NCCL. Aruntimeimage that also includes cuDNN is available. Some images may also include TensorRT.devel: Builds on theruntimeand 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.
NVIDIA Container Toolkit
The https://github.com/NVIDIA/nvidia-container-toolkit for Docker is required to run CUDA images.
For CUDA 10.0, nvidia-docker2 (v2.1.0) or greater is recommended. It is also recommended to use Docker 19.03.
How to report a problem
Read https://github.com/NVIDIA/nvidia-docker/wiki/Frequently-Asked-Questions to see if the problem has been encountered before.
After it has been determined the problem is not with the NVIDIA runtime, report an issue at the CUDA Container Image Issue Tracker.
Supported tags
Supported tags are updated to the latest CUDA, cuDNN, and TensorRT versions. These tags are also periodically updated to fix CVE vulnerabilities.
For a full list of supported tags, click here.
LATEST CUDA 13.2.0
Visit OpenSource @ Nvidia for the GPL sources of the packages contained in the CUDA base image layers.
ubuntu24.04 [arm64, x86_64]
13.2.0-cudnn-runtime-ubuntu24.04(13.2.0/ubuntu2404/runtime/cudnn/Dockerfile)13.2.0-runtime-ubuntu24.04(13.2.0/ubuntu2404/runtime/Dockerfile)13.2.0-cudnn-devel-ubuntu24.04(13.2.0/ubuntu2404/devel/cudnn/Dockerfile)13.2.0-devel-ubuntu24.04(13.2.0/ubuntu2404/devel/Dockerfile)13.2.0-base-ubuntu24.04(13.2.0/ubuntu2404/base/Dockerfile)
ubuntu22.04 [arm64, x86_64]
13.2.0-cudnn-runtime-ubuntu22.04(13.2.0/ubuntu2204/runtime/cudnn/Dockerfile)13.2.0-runtime-ubuntu22.04(13.2.0/ubuntu2204/runtime/Dockerfile)13.2.0-cudnn-devel-ubuntu22.04(13.2.0/ubuntu2204/devel/cudnn/Dockerfile)13.2.0-devel-ubuntu22.04(13.2.0/ubuntu2204/devel/Dockerfile)13.2.0-base-ubuntu22.04(13.2.0/ubuntu2204/base/Dockerfile)
ubi9 [arm64, x86_64]
13.2.0-cudnn-runtime-ubi9(13.2.0/ubi9/runtime/cudnn/Dockerfile)13.2.0-runtime-ubi9(13.2.0/ubi9/runtime/Dockerfile)13.2.0-cudnn-devel-ubi9(13.2.0/ubi9/devel/cudnn/Dockerfile)13.2.0-devel-ubi9(13.2.0/ubi9/devel/Dockerfile)13.2.0-base-ubi9(13.2.0/ubi9/base/Dockerfile)
ubi8 [arm64, x86_64]
13.2.0-cudnn-runtime-ubi8(13.2.0/ubi8/runtime/cudnn/Dockerfile)13.2.0-runtime-ubi8(13.2.0/ubi8/runtime/Dockerfile)13.2.0-cudnn-devel-ubi8(13.2.0/ubi8/devel/cudnn/Dockerfile)13.2.0-devel-ubi8(13.2.0/ubi8/devel/Dockerfile)13.2.0-base-ubi8(13.2.0/ubi8/base/Dockerfile)
ubi10 [arm64, x86_64]
13.2.0-cudnn-runtime-ubi10(13.2.0/ubi10/runtime/cudnn/Dockerfile)13.2.0-runtime-ubi10(13.2.0/ubi10/runtime/Dockerfile)13.2.0-cudnn-devel-ubi10(13.2.0/ubi10/devel/cudnn/Dockerfile)13.2.0-devel-ubi10(13.2.0/ubi10/devel/Dockerfile)13.2.0-base-ubi10(13.2.0/ubi10/base/Dockerfile)
suse16 [arm64, x86_64]
13.2.0-runtime-suse16(13.2.0/suse16/runtime/Dockerfile)13.2.0-devel-suse16(13.2.0/suse16/devel/Dockerfile)13.2.0-base-suse16(13.2.0/suse16/base/Dockerfile)
rockylinux9 [arm64, x86_64]
13.2.0-cudnn-runtime-rockylinux9(13.2.0/rockylinux9/runtime/cudnn/Dockerfile)13.2.0-runtime-rockylinux9(13.2.0/rockylinux9/runtime/Dockerfile)13.2.0-cudnn-devel-rockylinux9(13.2.0/rockylinux9/devel/cudnn/Dockerfile)13.2.0-devel-rockylinux9(13.2.0/rockylinux9/devel/Dockerfile)13.2.0-base-rockylinux9(13.2.0/rockylinux9/base/Dockerfile)
rockylinux8 [arm64, x86_64]
13.2.0-cudnn-runtime-rockylinux8(13.2.0/rockylinux8/runtime/cudnn/Dockerfile)13.2.0-runtime-rockylinux8(13.2.0/rockylinux8/runtime/Dockerfile)13.2.0-cudnn-devel-rockylinux8(13.2.0/rockylinux8/devel/cudnn/Dockerfile)13.2.0-devel-rockylinux8(13.2.0/rockylinux8/devel/Dockerfile)13.2.0-base-rockylinux8(13.2.0/rockylinux8/base/Dockerfile)
rockylinux10 [arm64, x86_64]
13.2.0-cudnn-runtime-rockylinux10(13.2.0/rockylinux10/runtime/cudnn/Dockerfile)13.2.0-runtime-rockylinux10(13.2.0/rockylinux10/runtime/Dockerfile)13.2.0-cudnn-devel-rockylinux10(13.2.0/rockylinux10/devel/cudnn/Dockerfile)13.2.0-devel-rockylinux10(13.2.0/rockylinux10/devel/Dockerfile)13.2.0-base-rockylinux10(13.2.0/rockylinux10/base/Dockerfile)
oraclelinux9 [arm64, x86_64]
13.2.0-cudnn-runtime-oraclelinux9(13.2.0/oraclelinux9/runtime/cudnn/Dockerfile)13.2.0-runtime-oraclelinux9(13.2.0/oraclelinux9/runtime/Dockerfile)13.2.0-cudnn-devel-oraclelinux9(13.2.0/oraclelinux9/devel/cudnn/Dockerfile)13.2.0-devel-oraclelinux9(13.2.0/oraclelinux9/devel/Dockerfile)13.2.0-base-oraclelinux9(13.2.0/oraclelinux9/base/Dockerfile)
oraclelinux8 [arm64, x86_64]
13.2.0-cudnn-runtime-oraclelinux8(13.2.0/oraclelinux8/runtime/cudnn/Dockerfile)13.2.0-runtime-oraclelinux8(13.2.0/oraclelinux8/runtime/Dockerfile)13.2.0-cudnn-devel-oraclelinux8(13.2.0/oraclelinux8/devel/cudnn/Dockerfile)13.2.0-devel-oraclelinux8(13.2.0/oraclelinux8/devel/Dockerfile)13.2.0-base-oraclelinux8(13.2.0/oraclelinux8/base/Dockerfile)
opensuse15 [x86_64]
13.2.0-cudnn-runtime-opensuse15(13.2.0/opensuse15/runtime/cudnn/Dockerfile)13.2.0-runtime-opensuse15(13.2.0/opensuse15/runtime/Dockerfile)13.2.0-cudnn-devel-opensuse15(13.2.0/opensuse15/devel/cudnn/Dockerfile)13.2.0-devel-opensuse15(13.2.0/opensuse15/devel/Dockerfile)13.2.0-base-opensuse15(13.2.0/opensuse15/base/Dockerfile)
azl3 [arm64, x86_64]
13.2.0-runtime-azl3(13.2.0/azl3/runtime/Dockerfile)13.2.0-devel-azl3(13.2.0/azl3/devel/Dockerfile)13.2.0-base-azl3(13.2.0/azl3/base/Dockerfile)
amzn2023 [arm64, x86_64]
13.2.0-runtime-amzn2023(13.2.0/amzn2023/runtime/Dockerfile)13.2.0-devel-amzn2023(13.2.0/amzn2023/devel/Dockerfile)13.2.0-base-amzn2023(13.2.0/amzn2023/base/Dockerfile)
Unsupported tags
A list of tags that are no longer supported can be found here
Source of this description
This Readme is located in the doc directory of the CUDA Container Image source repository. (history)
rocker/cuda
rocker
集成NVIDIA CUDA库的Rocker镜像，提供R语言环境下的GPU加速计算支持，基于rocker-org/rocker-versioned2项目构建。
10 次收藏5万+ 次下载
14 天前更新
okteto/cuda
okteto
暂无描述
172 次下载
1 年前更新
openeuler/cuda
openeuler
官方CUDA Docker镜像，基于openEuler构建，提供异构计算平台，通过分层API和高级库助力快速构建基于NVIDIA GPU的高性能计算应用和AI服务，免费使用且无用户速率限制。
650 次下载
7 个月前更新
vistart/cuda
vistart
基于NVIDIA官方Dockerfile构建的CUDA镜像，使用清华源，集成CUDA、CUDNN、NCCL、TensorRT等组件，支持多种Ubuntu版本，每周更新，提供Docker Hub和阿里云香港节点双存储。
17 次收藏5万+ 次下载
4 年前更新
gpuci/cuda
gpuci
用于填补空白并补充nvidia/cuda的CUDA镜像
5万+ 次下载
4 年前更新
tobycheese/cuda
tobycheese
基于Ubuntu 18.04的Docker镜像，集成CUDA 9.0和cuDNN 7，解决Nvidia未提供旧版CUDA与新版Ubuntu组合的问题，适用于依赖CUDA 9.0的应用场景。
2 次收藏1万+ 次下载
7 年前更新
轩辕镜像配置手册
探索更多轩辕镜像的使用方法，找到最适合您系统的配置方式
Docker 配置
登录仓库拉取
通过 Docker 登录认证访问私有仓库
专属域名拉取
无需登录使用专属域名
K8s Containerd
Kubernetes 集群配置 Containerd
K3s
K3s 轻量级 Kubernetes 镜像加速
Dev Containers
VS Code Dev Containers 配置
Podman
Podman 容器引擎配置
Singularity/Apptainer
HPC 科学计算容器配置
其他仓库配置
ghcr、Quay、nvcr 等镜像仓库
Harbor 镜像源配置
Harbor Proxy Repository 对接专属域名
Portainer 镜像源配置
Portainer Registries 加速拉取
Nexus 镜像源配置
Nexus3 Docker Proxy 内网缓存
系统配置
需要其他帮助？请查看我们的 常见问题Docker 镜像访问常见问题解答 或 提交工单
镜像拉取常见问题
使用与功能问题
错误码与失败问题
docker pull 提示 manifest unknown 怎么办？
manifest unknown
docker pull 提示 no matching manifest 怎么办？
no matching manifest（架构）
镜像已拉取完成，却提示 invalid tar header 或 failed to register layer 怎么办？
invalid tar header（解压）
Docker pull 时 HTTPS / TLS 证书验证失败怎么办？
TLS 证书失败
Docker pull 时 DNS 解析超时或连不上仓库怎么办？
DNS 超时
Docker 拉取出现 410 Gone 怎么办？
410 Gone 排查
出现 402 或「流量用尽」提示怎么办？
402 与流量用尽
Docker 拉取提示 UNAUTHORIZED（401）怎么办？
401 认证失败
遇到 429 Too Many Requests（请求太频繁）怎么办？
429 限流
docker login 提示 Cannot autolaunch D-Bus，还算登录成功吗？
D-Bus 凭证提示
为什么会出现「单层超过 20GB」或 413，无法加速拉取？
413 与超大单层
账号 / 计费 / 权限
用户好评
来自真实用户的反馈，见证轩辕镜像的优质服务
oldzhang
运维工程师
Linux服务器
5
"Docker访问体验非常流畅，大镜像也能快速完成下载。"