Unified Mirroring

ModelArts provides unified images of Arm+Ascend specifications, including MindSpore and PyTorch, which are used for development environments, model training, and service deployment. For details, see Unified Image List.

Unified Image List

ModelArts provides unified images of Arm+Ascend specifications, including MindSpore and PyTorch. You can use the images to develop environment, train models, and deploy services. For details, see Unified Image List.

**Table 1** MindSpore
Preset Image	Supported Processor	Applicable Scope
mindspore_2.2.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b	Ascend snt9b	Notebook, training, and inference deployment

**Table 2** PyTorch
Preset Image	Supported Processor	Applicable Scope
pytorch_2.1.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b	Ascend snt9b	Notebook, training, and inference deployment
pytorch_1.11.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b	Ascend snt9b	Notebook, training, and inference deployment

mindspore_2.2.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b

**Table 3** Introduction to mindspore_2.2.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b images
AI Engine	URL	Dependency
mindspore 2.2.0 + mindspore-lite 2.2.0 + Ascend CANN Toolkit 7.0.RC1	swr.<region>.myhuaweicloud.com/atelier/mindspore_2_2_ascend:mindspore_2.2.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b-20231107190844-50a1a83 For example: Hong Kong swr.ap-southeast-1.myhuaweicloud.com/atelier/mindspore_2_2_ascend:mindspore_2.2.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b-20231107190844-50a1a83	PyPI package	YUM package
		mindspore 2.2.0 ipykernel 6.7.0 ipython 8.17.2 jupyter-client 7.4.9 ma-cau 1.1.7 ma-cau-adapter 1.1.3 ma-cli 1.2.3 matplotlib 3.5.1 modelarts 1.4.20 moxing-framework 2.2.3.2c7f2141 numpy 1.22.0 pandas 1.2.5 pillow 10.0.1 pip 21.0.1 psutil 5.9.5 PyYAML 6.0.1 scipy 1.10.1 scikit-learn 1.0.2 tornado 6.3.3 mindinsight 2.2.0	cmake cpp curl ffmpeg g++ gcc git grep python3 rpm tar unzip wget zip

pytorch_2.1.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b

**Table 4** Introduction to pytorch_2.1.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b images
AI Engine	URL	Dependency
pytorch 2.1.0 + mindspore-lite 2.2.0 + Ascend CANN Toolkit 7.0.RC1	swr.<region>.myhuaweicloud.com/atelier/pytorch_2_1_ascend:pytorch_2.1.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b-20231107190844-50a1a83 For example: Hong Kong swr.ap-southeast-1.myhuaweicloud.com/atelier/pytorch_2_1_ascend:pytorch_2.1.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b-20231107190844-50a1a83	PyPI package	YUM package
		torch 2.1.0 apex 0.1-ascend-20231013 torch-npu 2.1.0rc1.post20231013 ipykernel 5.3.4 ipython 7.34.0 jupyter-client 7.3.4 ma-cau 1.1.6 ma-cau-adapter 1.1.3 ma-cli 1.2.2 matplotlib 3.5.1 modelarts 1.4.20 moxing-framework 2.2.3.2c7f2141 numpy 1.26.1 pandas 1.3.5 pillow 10.0.1 pip 21.0.1 psutil 5.9.5 PyYAML 6.0.1 scipy 1.10.1 scikit-learn 1.0.2 tornado 6.3.3	cmake cpp curl ffmpeg g++ gcc git grep python3 rpm tar unzip wget zip

pytorch_1.11.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b

**Table 5** Introduction to pytorch_1.11.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b images
AI Engine	URL	Dependency
pytorch 1.11 + mindspore-lite 2.2.0 + Ascend CANN Toolkit 7.0.RC1	swr.<region>.myhuaweicloud.com/atelier/pytorch_1_11_ascend:pytorch_1.11.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b-20231107190844-50a1a83 For example: Hong Kong swr.ap-southeast-1.myhuaweicloud.com/atelier/pytorch_1_11_ascend:pytorch_1.11.0-cann_7.0.1-py_3.9-euler_2.10.7-aarch64-snt9b-20231107190844-50a1a83	PyPI package	YUM package
		torch 1.11.0 torch-npu 1.11.0.post4-20231013 apex 0.1-ascend-20231013 ipykernel 6.7.0 ipython 8.17.2 jupyter-client 7.4.9 ma-cau 1.1.7 ma-cau-adapter 1.1.3 ma-cli 1.2.3 matplotlib 3.5.1 modelarts 1.4.20 moxing-framework 2.2.3.2c7f2141 numpy 1.26.1 pandas 1.3.5 pillow 10.0.1 pip 21.0.1 psutil 5.9.5 PyYAML 6.0.1 scipy 1.10.1 scikit-learn 1.0.2 tornado 6.3.3	cmake cpp curl ffmpeg g++ gcc git grep python3 rpm tar unzip wget zip