Help Center/ ModelArts/ Service Overview/ What Is ModelArts?
Updated on 2025-02-07 GMT+08:00
View PDF
Share

What Is ModelArts?

ModelArts is a one-stop development platform provided by Huawei Cloud. With large-volume data preprocessing, semi-automated data labeling, distributed training, automated model building, and on-demand model deployment across devices, edge devices, and cloud, ModelArts helps you quickly build and deploy models and efficiently manage the AI development lifecycle.

ModelArts covers all stages of AI development, including data processing, algorithm development, and model training and deployment. The underlying technologies of ModelArts support a wide range of heterogeneous computing resources, allowing you to flexibly select and use the resources that fit your needs. ModelArts supports popular open-source AI development frameworks such as TensorFlow, PyTorch, and MindSpore. ModelArts also allows you to use customized algorithm frameworks tailored to your needs.

Offerings

ModelArts offers several products.

Table 1 ModelArts offerings

Offering

Overview

Scenario

ModelArts Standard

ModelArts Standard is a one-stop platform for AI development. It offers a user-friendly console with integrated toolchains for ExeML, data management, development environments, model training, management, and deployment. This enables seamless management of the AI development lifecycle.

It is ideal for users who require an AI development platform.

ModelArts MaaS

ModelArts MaaS offers an end-to-end toolchain for foundation model production, along with Ascend computing resources and popular open-source models. It enables data production, model fine-tuning, prompt engineering, and application orchestration.

It is designed for users who need to develop production-ready models using a MaaS platform.

ModelArts Lite Server

ModelArts Lite Server offers cloud servers running on bare metal servers, accessible via EIPs.

It is designed for users who have built their own AI development platforms and require only computing power. It provides cost-effective AI computing power, mainstream AI development suites, and Huawei's acceleration plug-ins.

ModelArts Lite Cluster

ModelArts Lite Cluster provides direct access to Kubernetes APIs, enabling you to manage nodes and clusters within resource pools.

It is designed for users who have built their own AI development platforms and require only computing power. It requires basic knowledge of Kubernetes.

ModelArts Edge

ModelArts Edge simplifies the deployment and management of edge computing. It supports various types of edge devices and offers features such as model deployment, AI application and node management, resource pooling and load balancing, and application commercial use assurance. With ModelArts Edge, you can quickly build cost-effective AI solutions that leverage both edge and cloud resources.

It is suitable for edge deployment.

AI Gallery

AI Gallery offers a high-quality model development experience on Ascend Cloud and provides access to a wealth of community resources.

It is suitable for AI development and exploration.

Architecture

  • The computing power layer offers a full range of Ascend hardware, 10,000-card cluster management, and resource scheduling and management. It supports popular AI development, debugging, training, and inference frameworks.
  • The AI platform layer offers an end-to-end AI development toolchain, enabling you to develop and deploy models quickly and easily. It also enables efficient resource management and automatic fault recovery, speeding up AI model development, training, and deployment.
  • The AI development toolchain layer offers an end-to-end foundation model development toolchain, including high-quality open-source models and development suites that streamline the development process and reduce time to market.

Access Methods

ModelArts provides multiple access methods for different products.

  • Console

    ModelArts Standard can be accessed through the console, where you can perform various tasks such as ExeML, data management, model training, model management, and model deployment. The console allows for end-to-end AI development.

    ModelArts MaaS can be accessed through the console, where you can perform various tasks such as data production, model fine-tuning, prompt engineering, and application orchestration.

  • SDK

    ModelArts Standard can be called through SDKs to be integrated into a third-party system for secondary development. ModelArts SDKs encapsulate ModelArts Standard RESTful APIs in Python to simplify development. For details about the SDKs and operations, see the ModelArts SDK Reference.

    Additionally, you can directly call the ModelArts SDKs when writing code in ModelArts Standard Notebook.

  • API

    If you want to integrate ModelArts Standard into a third-party system for secondary development, use APIs to access ModelArts..

  • Cloud native

    ModelArts Lite Server allows you to use EIPs to access your cloud servers. For details, see ModelArts Lite Server User Guide.

    ModelArts Lite Cluster allows you to use native K8s APIs to manage your clusters. For details, see ModelArts Lite Cluster User Guide.