Rate This Document
Findability
Accuracy
Completeness
Readability

Environment Requirements

This document uses openEuler 24.03 LTS SP3 + PyTorch 2.11.0 + CUDA 13.0 as an example. In addition, this section describes other environments supported by PyTorch installation on Kunpeng servers.

Supported Versions

Table 1 Supported versions

OS

Python

PyTorch

openEuler 24.03 LTS SP3

Python 3.11.6

PyTorch 2.11.x (PyTorch 2.11.0 is recommended.)

openEuler 22.03 LTS SP4

Python 3.9.9

PyTorch <= 2.8.x (PyTorch 2.8.0 is recommended.)

PyTorch 2.11.0 requires Python 3.10 or later. The default Python version of openEuler 22.03 LTS SP4 is 3.9.9, which does not meet the version requirements. You are advised to use PyTorch 2.8.0 that supports Python 3.9 on openEuler 22.03 LTS SP4.

Dependent Component Requirements

Table 2 Version requirements for dependent components

Component

Version

python3

3.11.6-20.oe2403sp3

python3-pip

23.3.1-7.oe2403sp3

pip

26.1.1

setuptools

82.0.1

wheel

0.47.0

PyTorch

2.11.0

CUDA Toolkit

13.0

To use the GPU capability, the system must provide an accessible NVIDIA GPU device and the NVIDIA driver user-mode library that matches CUDA 13.0. During the installation verification, ensure that PyTorch is a CUDA version and performs GPU tensor computation.