Jupyter Notebook for interactive programming development
Documentation
Technical documentation
Learn about the technical information and best practices about Platform for AI.
Platform for AI of Alibaba Cloud is a machine learning or deep learning engineering platform intended for enterprises and developers. It provides easy-to-use, cost-effective, high-performance, and easy-to-scale plug-ins that can be applied to various industry scenarios. With over 140 built-in optimization algorithms, Platform for AI provides whole-process AI engineering capabilities including data labeling (PAI-iTAG), model building (PAI-Designer and PAI-DSW), model training (PAI-DLC), compilation optimization, and inference deployment (PAI-EAS).
What's New in Alibaba Cloud Generative AI
AI Engineering Platform
Preparing Data
In the data preparation phase, PAI-iTAG provides intelligent data labeling services. PAI-iTAG supports different types of data, such as image, text, video, and audio, supports multimodal data labeling, and provides various labeling content components and topic components. You can use predefined labeling templates provided by PAI-iTAG or customize your own labeling templates. It also provides fully managed data labeling services that are outsourced.
Model Development
In the model development phase, you can use PAI-Designer and PAI-DSW to develop models.
PAI-Designer
Supports different types of hardware resources, including CPUs and GPUs, and features high throughput and low latency. It allows you to deploy large-scale complex models with a few clicks and perform elastic scale-ins and scale-outs in real time. It also provides a comprehensive O&M and monitoring system.
PAI-DSW
PAI-Blade provides universal optimization technologies that can be applied to different business scenarios. You can run your models at optimal inference performance and greatly improve the model QPS by using PAI-Blade to implement joint optimization.
Model training
In the model training phase, you can use PAI-DLC, which is a one-stop platform for cloud-native deep learning and training, to train models. It is compatible with predefined algorithm frameworks, allows you to customize algorithm frameworks, and supports mega-scale task execution for distributed deep learning. PAI-DLC features high flexibility, high stability, high performance, and ease of use.
Model Deployment
In the model deployment phase, you can use PAI-EAS to perform online inference and use PAI-Blade to implement inference optimization.
PAI-EAS Elastic Inference Service Platform
Supports different types of hardware resources, including CPUs and GPUs, and features high throughput and low latency. It allows you to deploy large-scale complex models with a few clicks and perform elastic scale-ins and scale-outs in real time. It also provides a comprehensive O&M and monitoring system.
PAI-Blade Universal Inference Accelerator
PAI-Blade provides universal optimization technologies that can be applied to different business scenarios. You can run your models at optimal inference performance and greatly improve the model QPS by using PAI-Blade to implement joint optimization.
AI Asset Management
Machine Learning Platform for AI provides full lifecycle management for major data assets used for AI-based production, such as models, datasets, and images, as well as output data. In addition, it allows you to share data assets with other users, analyze the training results of different models, and backtrack issues and exceptions. Machine Learning Platform for AI helps you improve performance and save costs throughout the process of AI development and application.
Specifications
Diverse purchase methods to meet the requirements of AI and machine learning at different stages.
PAI-DSW
PAI-DSW is an interactive programming development tool of Machine Learning Platform for AI and can be used by developers of different levels.
AI programming code development tool
-
![]()
Jupyter Notebook for interactive programming development
-
![]()
Provides deeply optimized TensorFlow and supports open source framework
-
![]()
Supports hybrid scheduling of CPUs or GPUs
PAI-Designer
PAI-Designer is a codeless development tool that provides classic machine learning algorithm components, covering scenarios like regression, classification, clustering, and text analysis.
Codeless development tool
-
![]()
Over 140 algorithm components
-
![]()
Drag-and-drop algorithm without the need to write code
-
![]()
Adjusts parameters more conveniently
PAI-EAS
PAI-EAS is a model deployment tool. It allows you to quickly deploy your algorithm models and perform online debugging.
Model deployment and debugging services
-
![]()
Deploys complex models with a few clicks
-
![]()
Creates, updates, stops, starts, scales out, and scales in model services
-
![]()
Supports online debugging of models
Documentation and Learning Path
Upgraded Support For You
1 on 1 Presale Consultation, 24/7 Technical Support, Faster Response, and More Free Tickets.
1 on 1 Presale Consultation
Consulting by experienced cloud experts.Learn More
24/7 Technical Support
Extended service time from 10 hours 5 days a week to 24/7. Learn More
6 Free Tickets per Quarter
The number of free tickets doubled from 3 to 6 per quarter. Learn More
Faster Response
Shorten after-sale response time from 36 hours to 18 hours. Learn More
