Installing CANN (Online Installation Using pip)

Prerequisites

Before installation, ensure that the Python environment and pip3 are available. CANN currently supports Python versions 3.7.x to 3.14.x. If this requirement is not met, run the following commands to install them:

1

sudo apt-get install -y python3 python3-pip

1

sudo yum install -y python3 python3-pip

If the installation fails, the version does not meet the requirements, or the dynamic library libpython3.x.so is not included, perform the operations in Compiling and Installing Python.

Configuring the Ascend Repository

Configure the Ascend repository. Otherwise, the installation will fail.

pip config set global.extra-index-url https://ascend.devcloud.huaweicloud.com/cann/pypi/simple
pip config set global.trusted-host ascend.devcloud.huaweicloud.com

Installing the Toolkit

The pip-installed CANN Toolkit is intended only for framework-side training and inference workloads. If you need to install the Toolkit for development and debugging (such as development and build of operators, applications, and models), select another installation method.

Before installing the Toolkit, make sure the installation directory has more than 10 GB of free space. If not, clear some space or choose another directory.

1. Run the installation command:

   1	pip install ascend-cann-toolkit==9.1.0

   The software package is installed in the <Python path>/site-packages/ascend directory. You can run the pip show ascend-cann-toolkit command to query the specific path of the software package.

2. Set the environment variable. The following uses installation path ${HOME}/lib/python3.14/site-packages/ascend as an example.

   1	source ${HOME}/lib/python3.14/site-packages/ascend/cann-9.1.0/set_env.sh

   The preceding environment variables take effect only in the current window. You can add the preceding command to the environment variable configuration file (for example, the .bashrc file) as required.

Installing ops

The CANN ops package integrates a comprehensive suite of libraries, including the operator foundational framework, operator libraries (covering math, nn, cv, transformer, and ras), TBE operators, HCCL, HIXL, and the DVPP library. Specifically, it provides dynamic and static library files for single-operator API execution (e.g., aclnn-class APIs), operator source code, and kernel binaries, ultimately optimizing overall execution performance in high-performance computing (HPC) workloads.

Before installing ops, ensure that the Toolkit of a compatible version has been installed in the same path. Select the ops software package corresponding to the running device. Installing ops for multiple chips in the same path is not supported. To work with different chips, install the Toolkit and the related ops in separate paths for multi-chip development and deployment.

1

pip install ascend-cann-950-ops==9.1.0

1

pip install ascend-cann-a3-ops==9.1.0

1

pip install ascend-cann-910b-ops==9.1.0

1

pip install ascend-cann-910-ops==9.1.0

1

pip install ascend-cann-310p-ops==9.1.0

1

pip install ascend-cann-310b-ops==9.1.0