MindSpore Quantum Installation

Confirming System Environment Information

Before you begin, please confirm that your software and hardware environment meets the following requirements:

Python Version: Please ensure your Python environment is version 3.9, 3.10, or 3.11. You can check your version with the python --version command.

Version Compatibility

Before installing MindQuantum, you must first install the corresponding version of MindSpore based on the compatibility table below. A version mismatch may cause installation or runtime issues.

MindQuantum Version	Required MindSpore Version
0.11.0	MindSpore >= 2.2.0 (2.7.0rc1 is recommended)

Please refer to the MindSpore Installation Guide to install the recommended version of MindSpore.

Installation Methods

You can install MindQuantum either by pip or from source code.

Installing by pip

1. Installing the Latest Stable Version

You can install the latest stable version of MindQuantum from the official PyPI source.

pip install mindquantum

2. Installing a Specific Version

You can also download a specific version's whl package from the MindSpore community and install it manually.

Please select the appropriate mindquantum-0.11.0 version from the table below based on your system environment.

Operating System	Hardware Platform	Python Version	MindQuantum whl Package	SHA-256 Checksum File
Linux	x86_64 (CPU/GPU)	3.9	mindquantum-0.11.0-cp39-cp39-linux_x86_64.whl	sha256
		3.10	mindquantum-0.11.0-cp310-cp310-linux_x86_64.whl	sha256
		3.11	mindquantum-0.11.0-cp311-cp311-linux_x86_64.whl	sha256
	aarch64 (CPU)	3.9	mindquantum-0.11.0-cp39-cp39-linux_aarch64.whl	sha256
		3.10	mindquantum-0.11.0-cp310-cp310-linux_aarch64.whl	sha256
		3.11	mindquantum-0.11.0-cp311-cp311-linux_aarch64.whl	sha256
Windows	x86_64 (CPU)	3.9	mindquantum-0.11.0-cp39-cp39-win_amd64.whl	sha256
		3.10	mindquantum-0.11.0-cp310-cp310-win_amd64.whl	sha256
		3.11	mindquantum-0.11.0-cp311-cp311-win_amd64.whl	sha256
macOS	x86_64 (CPU)	3.9	mindquantum-0.11.0-cp39-cp39-macosx_10_15_x86_64.whl	sha256
		3.10	mindquantum-0.11.0-cp310-cp310-macosx_10_15_x86_64.whl	sha256
		3.11	mindquantum-0.11.0-cp311-cp311-macosx_10_15_x86_64.whl	sha256
	arm64 (CPU)	3.9	mindquantum-0.11.0-cp39-cp39-macosx_11_0_arm64.whl	sha256
		3.10	mindquantum-0.11.0-cp310-cp310-macosx_11_0_arm64.whl	sha256
		3.11	mindquantum-0.11.0-cp311-cp311-macosx_11_0_arm64.whl	sha256

About the Linux x86_64 package: This platform provides a unified CPU/GPU version. After installation, if your device has a compatible NVIDIA GPU and a CUDA 11.1 or higher environment, GPU-related features will be available.

Installation Commands

You can install directly via URL using pip, or download the whl package and its sha256 checksum file to install locally.

# Method 1: Install directly from URL (e.g., for Linux x86_64, Python 3.9)
pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.7.0rc1/MindQuantum/gpu/x86_64/cuda-11.1/mindquantum-0.11.0-cp39-cp39-linux_x86_64.whl

Installation and Verification from Local Download

Download Files: Download the whl package and its corresponding .whl.sha256 checksum file from the table above.
Verify the Hash (optional)

First, compute the hash value of your downloaded whl file (replace your_downloaded_file.whl with your actual filename):
```
sha256sum your_downloaded_file.whl
```
Next, display the official hash value recorded in the .sha256 file:
```
cat your_downloaded_file.whl.sha256
```
Manually compare the hash strings from the two outputs. If they are identical, the file is correct.

Note: Please be aware that the filename recorded in the .whl.sha256 file may not exactly match the actual filename of the downloaded whl package. This can happen because the build system sometimes standardizes filenames for broader compatibility (e.g., renaming manylinux_2_27_x86_64 to linux_x86_64). Please use the hash value as the authoritative reference.

Install the Package: After successful verification, install the package using pip:

# Replace your_downloaded_file.whl with your actual filename
pip install your_downloaded_file.whl

If you have a poor network connection, you can try using the Huawei Cloud mirror to speed up the download: pip install -i https://repo.huaweicloud.com/repository/pypi/simple mindquantum.

Installing by Source Code

Download Source Code from Gitee

cd ~
git clone -b r0.11 https://gitee.com/mindspore/mindquantum.git

Compiling MindSpore Quantum

cd ~/mindquantum
python setup.py install --user

Verifying Successful Installation

If no error message such as No module named 'mindquantum' is reported when you run the following command, the installation is successful:

python -c 'import mindquantum'

Installing with Docker

Mac or Windows users can install MindSpore Quantum through Docker. Please refer to the Docker installation guide.