Help Center> ModelArts> Preparations> (Optional) Buying a Package
View PDF

(Optional) Buying a Package

ModelArts Packages

ModelArts allows you to pay as you go or buy packages. The fee varies according to the selected resources. You can choose either of the following packages to meet your service requirements:

  • AI Development Lifecycle: Dedicated for developers with AI development experience. It supports machine learning and deep learning algorithm development and deployment, including data processing, model development, training, and management, and service deployment. The billing items include model development environments (Notebook), model training (training and virtualization jobs), and service deployment (real-time services).
  • ExeML: Dedicated for developers with few AI development capabilities. It supports data labeling, automatic design, optimization, and model training, as well as service deployment, delivering code-free AI development. This package applies only to training and deployment of ExeML jobs. The billing items include ExeML projects (training jobs and service deployment).

AI Development Lifecycle

The computing resources involved in the AI development lifecycle are classified into Compute CPU (2U) instance, Compute CPU (8U) instance, Compute GPU (P100) instance, and Compute GPU (V100 NVLINK_32G) instance. For details about the specifications and supported functions, see Table 1.

Table 1 Computing resources involved in AI Development Lifecycle

Resource Name

Required Duration

Supported Function

Validity Period

Compute CPU (2U) instance
  • 300 hours
  • 1,000 hours
  • DevEnviron-Notebook
  • Model training-Training jobs
  • Model training-Visualization jobs
  • Service deployment-Real-time services

1 year

Compute CPU (8U) instance
  • 300 hours
  • 1,000 hours
  • DevEnviron-Notebook

Compute GPU (P100) instance
  • 300 hours
  • 1,000 hours
  • DevEnviron-Notebook
  • Model training-Training jobs

Compute GPU (V100 NVLINK 32_GB) instance
  • 300 hours
  • 1,000 hours
  • DevEnviron-Notebook
  • Model training-Training jobs

ExeML

The computing resources involved in ExeML are classified into Compute-intensive (P1) instance, Compute-intensive (P2) instance, and Compute-intensive (P3) instance. For details about the specifications and supported functions, see Table 2.

Table 2 Computing resources involved in ExeML

Resource Name

Required Duration

Supported Function

Validity Period

Compute-intensive (P1) instance
  • 300 hours
  • 1,000 hours

ExeML (training jobs)

1 year

Compute-intensive (P2) instance
  • 300 hours
  • 1,000 hours

ExeML (GPU-based deployment)

Compute-intensive (P3) instance
  • 300 hours
  • 1,000 hours

ExeML (CPU-based deployment)

Buying a Package

  1. Log in to the ModelArts management console and click Buy Package on the right. The Buy Package page is displayed.
  2. Select AI Development Lifecycle or ExeML based on service requirements.
  3. Select a package and click Next. The specifications confirmation page is displayed.
  4. After confirming the specifications, click Pay Now, and then complete the payment on the payment page.
    • During billing, the resource quota in a package is deducted first. After the quota is exhausted, pay-per-use rates apply. The quota of a package defines the available resources within one year from the date you bought the package. The validity period of a package is one year.
    • The package you bought cannot be unsubscribed.