Help Center/ ModelArts/ Best Practices/ ModelArts Best Practices
Updated on 2026-09-03 GMT+08:00
View PDF
Share

ModelArts Best Practices

This document provides ModelArts samples concerning a variety of scenarios and AI engines to help you quickly understand the process and operations of using ModelArts for AI development.

All practical case studies provided in this document are intended solely for scenarios such as function validation and prototype testing. Given the stringent requirements of production environments for high availability, security compliance, observability, and operations management, do not use the case studies in this document directly in production environments.

LLM Training and Inference

Sample

Scenario

Description

Deploying a Model from the vLLM-Ascend Open-Source Community on the ModelArts Inference Platform

Inference deployment

Describes how to deploy models obtained from the vLLM-Ascend open-source community on the ModelArts inference platform.

Video Generation Model Training and Inference

Sample

Scenario

Description

Inference Guide for Wan2.1, Wan2.2, HunyuanVideo, and CogVideo Series Models Adapted to NPU via ModelArts Lite Server (6.5.911)

Wan series model inference

Describes the inference process of Wan series models based on ModelArts Lite Server. The PyTorch framework and Ascend NPUs are used for inference.

ModelArts Development Environment Cases

Table 1 ModelArts samples

Sample

Function

Scenario

Description

Migrating the Conda Environment on a Notebook Instance to an SFS Disk

Environment migration

Development environments

Describes how to migrate the Conda environment of a notebook instance to an SFS disk.

ModelArts Inference Deployment

Table 2 Inference deployment samples

Sample

Scenario

Description

Accessing a Real-Time Service Through a Dedicated API Gateway, WAF, VPC, and ELB

Inference deployment

Describes how to use Dedicated API Gateway (APIG), Web Application Firewall (WAF), Virtual Private Cloud (VPC), and Elastic Load Balance (ELB) to access ModelArts real-time services over the intranet.