Table of contents
CUDA and Pytorch version correspondence table
Installation package download link
Preface
Overview of this article: Downloading CUDA from the official website requires scientific Internet access, and downloading cudnn requires an NVIDIA account, which is very troublesome. Therefore, the author specially downloaded 4 mainstream versions of CUDA and their corresponding CUDNN from the NVIDIA official website and uploaded them to Baidu Cloud Network Disk for everyone to download.
Author's introduction: The author is an artificial intelligence alchemist. Currently, the main research direction in the laboratory is generative models. He also has some knowledge of other directions. He hopes to communicate with friends who are also interested in artificial intelligence on the CSDN platform. Share and make progress together. Thank you everyone~~~
如果你觉得这篇文章对您有帮助，麻烦点赞、收藏或者评论一下，这是对作者工作的肯定和鼓励。
If you want other versions of CUDA or CUDNN, please reply in the comment area, and I will upload it to Baidu Cloud Disk as soon as possible.
CUDA and Pytorch version correspondence table
| CUDA version | Corresponding to CUDNN version | Supported pytorch versions |
| CUDA11.6 | CUDNN8.9.4 | pytoch1.13、1.12 |
| CUDA11.3 | CUDNN8.9.4 | pytorch1.12、1.11、1.10、1.9 |
| CUDA10.2 | CUDNN8.7.0 | pytorch1.10、1.9、1.8、1.7、1.6、1.5 |
| CUDA9.0 | CUDNN7.6.5 | pytorch1.1、1.0 |
If you want to know the CUDA version supported by your graphics card and the CUDA installation process, you can read my previous article: CUDA Installation Tutorial
Installation package download link
Link: https://pan.baidu.com/s/1vcb8peRdkls_AERXVXsleg?pwd=o34r
Extraction code: o34r
-- Sharing from Baidu Netdisk Super Member V4
Epilogue
如果您觉得这篇文章对您有帮忙，请点赞、收藏。您的点赞是对作者工作的肯定和鼓励，这对作者来说真的非常重要。如果您对文章内容有任何疑惑和建议，欢迎在评论区里面进行评论，我将第一时间进行回复。