cuda11.7支持的gpu
时间: 2023-09-25 08:09:39 浏览: 103
CUDA 11.7 支持以下 NVIDIA GPU:
1. Ampere 架构:A100, A30
2. Volta 架构:V100, V100S, V100 SMX2, V100-PCIE-32GB, GV100, GV100GL
3. Turing 架构:T4, TU102, TU104, TU106, TU116, TU117
4. Pascal 架构:GP100, GP102, GP104, GP106, GP107, GP108
5. Maxwell 架构:GM100, GM107, GM108, GM200, GM204, GM206, GM107M, GM108M, GM204M, GM206M
请注意,不是所有的 GPU 都支持 CUDA 11.7。要查看您的 GPU 是否支持 CUDA 11.7,请参考官方文档:https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html。
相关问题
cuda11.7安装gpu版pytorch
可以直接回答,以下是安装步骤:
1. 确认自己的显卡型号,并下载对应的CUDA11.7版本安装包。
2. 安装CUDA11.7,注意选择“Custom”安装方式,不要选择“Express”模式。
3. 安装cuDNN,需要先注册NVIDIA账号,然后下载对应版本的cuDNN压缩包,解压后复制到CUDA的安装目录中。
4. 确认自己的python版本,下载对应的pytorch安装包,安装时选择“Anaconda”或“Miniconda”环境。
5. 安装完成后,可以通过命令行或IDE测试是否安装成功。
cuda11.7对应的gpu版本
CUDA 11.7 支持的 GPU 版本包括:
- NVIDIA A100
- NVIDIA A40
- NVIDIA A30
- NVIDIA A10
- NVIDIA A16
- NVIDIA A16X
- NVIDIA A100-SXM4-80GB
- NVIDIA A100-SXM4-40GB
- NVIDIA A100-PCIE-80GB
- NVIDIA A100-PCIE-40GB
- NVIDIA A100-MIG-GPU-40GB
- NVIDIA A100-MIG-GPU-80GB
- NVIDIA A100-TENSOR-CORE-GPU
- NVIDIA A100-TENSOR-CORE-GPU-SXM4-40GB
- NVIDIA A100-TENSOR-CORE-GPU-SXM4-80GB
- NVIDIA T4
- NVIDIA V100
- NVIDIA P100
- NVIDIA K80
- NVIDIA M40
- NVIDIA M4
- NVIDIA A30-PCIe
- NVIDIA A10-PCIe
- NVIDIA T4-PCIe
- NVIDIA T4-OEM
- NVIDIA T4-Enterprise-Server
- NVIDIA T4-Server
- NVIDIA T4-Workstation
- NVIDIA V100-PCIe
- NVIDIA V100-SXM2-16GB
- NVIDIA V100-SXM2-32GB
- NVIDIA V100-SXM2-32GB-HBM2
- NVIDIA V100-SXM2-16GB-HBM2
- NVIDIA V100-SXM3-40GB
- NVIDIA V100-SXM3-80GB
- NVIDIA V100-SXM4-40GB
- NVIDIA V100-SXM4-80GB
- NVIDIA V100-PCIE-16GB
- NVIDIA V100-PCIE-32GB
- NVIDIA V100-DGX-1
- NVIDIA V100-DGX-2
- NVIDIA P40
- NVIDIA P4
- NVIDIA K80-PCIe
- NVIDIA K80-MX
- NVIDIA K80-DUO
- NVIDIA K40c
- NVIDIA K40m
- NVIDIA K40s
- NVIDIA K40st
- NVIDIA K40t
- NVIDIA GRID K2
- NVIDIA GRID K520
- NVIDIA GRID K1
- NVIDIA Tesla M2090
- NVIDIA Tesla M2075
- NVIDIA Tesla M2050
- NVIDIA Tesla M10
- NVIDIA Tesla K80
- NVIDIA Tesla K520
- NVIDIA Tesla K40c
- NVIDIA Tesla K40m
- NVIDIA Tesla K40s
- NVIDIA Tesla K40st
- NVIDIA Tesla K40t
- NVIDIA Tesla K20Xm
- NVIDIA Tesla K20m
- NVIDIA Tesla K10
- NVIDIA Tesla C2075
- NVIDIA Tesla C2070
- NVIDIA Tesla C2050
- NVIDIA Tesla C1060