报错解决之GeForce RTX *** with CUDA capability sm_86 is not compatible with the current PyTorch...

最新推荐文章于 2025-03-08 20:21:54 发布

哈哈哈哈哈嗝哈哈哈

最新推荐文章于 2025-03-08 20:21:54 发布

阅读量1.2w

点赞数 10

分类专栏： Debug of DL PyTorch 文章标签： pytorch 深度学习神经网络

本文链接：https://blog.csdn.net/weixin_43051346/article/details/123003905

版权

PyTorch 同时被 2 个专栏收录

13 篇文章

订阅专栏

Debug of DL

10 篇文章

订阅专栏

本文详细分析了在使用PyTorch进行深度学习时遇到的CUDA兼容性错误，解释了CUDA能力与版本之间的关系，并提供了解决方案——更新CUDA版本以匹配PyTorch。通过调整CUDA版本，确保显卡算力与支持的CUDA版本相匹配，可以避免此类错误并顺利进行模型训练。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

报错信息

深度学习神经网络模型训练的时候，可能会报出如下错误：

GeForce RTX *** with CUDA capability sm_86 is not compatible with the current PyTorch installation. 
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 compute_37.

问题分析

报错信息中的CUDA capability sm_86意思是算力8.6，即该RTX型号的显卡算力是8.6，但是当前的PyTorch依赖的CUDA版本支持的算力只有3.7、5.0、6.0、6.1、7.0、7.5。

经查阅资料¹，得到下述说明：

A CUDA application binary (with one or more GPU kernels) can contain the compiled GPU code in two forms, binary cubin objects and forward-compatible PTX assembly for each kernel. 
Both cubin and PTX are generated for a certain target compute capability. 
A cubin generated for a certain compute capability is supported to run on any GPU with the same major revision and same or higher minor revision of compute capability. 
For example, a cubin generated for compute capability 7.0 is supported to run on a GPU with compute capability 7.5, however a cubin generated for compute capability 7.5 is not supported to run on a GPU with compute capability 7.0, and a cubin generated with compute capability 7.x is not supported to run on a GPU with compute capability 8.x.