Installing CANN (Online Installation Using apt-get)

This section describes how to quickly install the CANN software using apt-get.

Dependency Installation

Before the installation, ensure that the Python environment and pip3 are available. Currently, CANN supports Python 3.7.x to 3.13.x. If the Python version does not meet the requirements, install it.

sudo apt-get 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.

wget https://ascend.devcloud.huaweicloud.com/cann/debian/cann-keyring_1.0.0_all.deb
sudo dpkg -i cann-keyring_1.0.0_all.deb
sudo apt-get update

Installing the Toolkit Development Kit

The CANN Toolkit development kit is installed in the training, inference, and development and debugging scenarios. It is used for training and inference services, model conversion, and development and compilation of operators, applications, and models.

Ensure that the available space of the installation directory is greater than 10 GB. If this requirement is not met, clear the space or change the installation directory.
Both Toolkit and ops must be installed in CANN 8.5.0.

Run the installation command.
1
sudo apt-get install ascend-cann-toolkit=8.5.0
During online installation using apt-get, the Toolkit can be installed only in the default path /usr/local/Ascend directory.
Configure environment variables.
1
source /usr/local/Ascend/cann/set_env.sh
The preceding environment variables take effect only in the current window. You can add the preceding commands to the environment variable configuration file (for example, the .bashrc file) as required.

Installing the ops Operator Package

Before installing the ops operator package, install Toolkit of the matching version in the same path. Currently, the ops operator packages of multiple chips cannot be installed in the same path. You can install Toolkit and ops of different chips in different paths to meet the development and deployment requirements in a multi-chip environment.

Table 1 Installing the ops Operator Package

Product Type

Installation Command

Atlas A3 series

sudo apt-get install ascend-cann-a3-ops=8.5.0

Atlas A2 series

sudo apt-get install ascend-cann-910b-ops=8.5.0

Atlas training series

sudo apt-get install ascend-cann-910-ops=8.5.0

Atlas inference series

sudo apt-get install ascend-cann-310p-ops=8.5.0

Atlas 200I/500 A2 inference series

sudo apt-get install ascend-cann-310b-ops=8.5.0

(Optional) Installing the NNAL Neural Network Acceleration Library

The NNAL neural network acceleration library provides the Ascend Transformer Boost (ATB) and AscendSiPBoost (SiP) signal processing acceleration libraries.

Before installing an acceleration library, install Toolkit of the matching version and configure environment variables.

Run the installation command.
1
sudo apt-get install ascend-cann-nnal=8.5.0
During online installation using apt-get, the Toolkit can be installed only in the default path /usr/local/Ascend directory.
Configure environment variables.
- ATB acceleration library:
  1
  source /usr/local/Ascend/nnal/atb/set_env.sh
- SiP acceleration library:
  1
  source /usr/local/Ascend/nnal/asdsip/set_env.sh
The preceding environment variables take effect only in the current window. You can add the preceding commands to the environment variable configuration file (for example, the .bashrc file) as required.