Data Agent is a data intelligence product built by Alibaba Cloud ApsaraDB based on Data+AI intelligence integration and Agentic AI technology trends. It covers the full lifecycle of data generation, storage, processing, administration, and analysis, with attributes such as autonomous planning, intelligent execution, and iterative optimization, enabling end-to-end automation from requirement analysis and data exploration to result delivery. Business users can obtain professional-level analysis results without mastering complex query languages or data analytics skills, while ensuring data accuracy and security.
Data Agent Enterprise Edition builds on Data Agent Personal Edition's basic data source connectivity, memory management, and custom MCP configuration. It also supports common databases such as on-premises, third-party cloud, and ECS self-managed databases, including MySQL, PostgreSQL, Hologres, MaxCompute, and 40+ enterprise-level data sources. It provides more professional, flexible, and controllable enterprise-level capabilities such as product integration, collaboration management, custom agents, and periodic runs.
Data Agent Enterprise offers high concurrency, enterprise-grade integration, and powerful collaboration. Click here to get started today!
Leveraging DMS's cross-cloud, multi-source data source management capabilities, DMS supports intelligent fusion of structured and unstructured data, covering the end-to-end flow from data collection, processing, and analytics to application.
Automatically parses business rules, breaks down complex requirements into executable jobs, builds a tiering processing frame, generates customized analysis suggestions, and outputs action solutions.
Starts an independent, dedicated data compute environment for each account to complete all data analytics jobs planned by the agent. The compute environment is owned by the user and connected to the enterprise VPC network, keeping data within the domain for enhanced security.
Continuously accumulate knowledge from business interactions, dynamically update enterprise-specific knowledge bases, and adapt to changing business requirements through continuous optimization.
Upload local CSV and Excel files or integrate with enterprise data assets such as RDS, PolarDB, and AnalyticDB. While ensuring enterprise data security, you can quickly specify the data to analyze. Data Agent explores and analyzes the specified data in detail, leveraging the cross-cloud, multi-source data source management capabilities that DMS has built over the years.
After you specify all the data and submit your analysis requirements, Data Agent starts working as your dedicated data analyst. Through in-depth understanding of requirements and deep insight into data, the agent invokes different tools based on job complexity to intelligently generate executable analysis schedules. You can also supplement or modify the schedule as needed.
You can confidently use reports generated by Data Agent. It completes analysis as scheduled by generating Python code for data computation, rather than directly outputting results by predicting the next token through an LLM. The entire computation procedure — including code generation, execution, iteration, and feedback at each step — is rendered for you to review every detail.
Finally, Data Agent outputs a well-structured, interactive data analytics report with visualization. The system automatically detects core business insights through intelligent algorithms and highlights them with annotations. You can interact with charts in the report to obtain multi-dimension analysis results in real time, facilitating decision-making and business optimization.
Data Agent is an enterprise data intelligence assistant based on a large language model. It offers three editions: Free Edition, Personal Edition, and Enterprise Edition, designed to meet individual users' basic and advanced requirements, as well as enterprise scenarios such as multi-user collaboration, security management, and private deployment. It supports data query, analysis, and processing through natural language conversations without writing code, empowering users at all levels to efficiently achieve data-driven decision-making.
Only Data Agent Enterprise Edition supports common databases such as on-premises, other-cloud, and ECS self-managed databases, including more than 40 enterprise-level data sources such as MySQL, PostgreSQL, Hologres, and MaxCompute.
• Free Edition: To help individual users experience and evaluate core features for the first time, the Free Edition is available for direct use after logon, but is subject to a daily 30-minute usage limit and concurrent usage limitations.
• Personal Edition: Provides developers and analysts with unlimited usage duration and concurrency. Purchased by seat to improve work efficiency.
• Enterprise Edition: Designed for teams and enterprises. Purchased through resource plans, it provides high concurrency and enterprise-level integration and collaboration capabilities, facilitating intelligent data integration for internal systems.
| Attribute description | Free Edition | Personal Edition | Enterprise Edition |
|---|---|---|---|
| Agent working hours | 30 minutes per day (Purged at 00:00 daily, non-cumulative) | No limitations | Unlimited |
| Task concurrency (number of concurrently executed jobs) | 3 | Each seat supports 3 jobs in parallel | The degree of parallelism for jobs is determined by the LLM concurrency you purchased. Jobs will be queued once the concurrency limitations are exceeded. |
| Basic data source (Local file, RDS, PolarDB, AnalyticDB for MySQL) | ✅ | ✅ | ✅ |
| Enterprise-level data sources (More than 40 types managed by DMS, such as Hologres and MaxCompute. For the complete list, see Databases supported by DMS) | ❌ | ❌ | ✅ |
| Product Integration (frontend IFrame embedding, OpenAPI integration) | ❌ | ❌ | ✅ |
| Memory management (Data Agent can automatically generate long-term memory based on user input and behavior, getting smarter the more you use it) | ✅ | ✅ | ✅ |
| Collaboration management (supports creating spaces for multi-user collaboration) | ❌ | ❌ | ✅ |
| Custom Agent (supports configuring a custom template for the Agent) | ❌ | ❌ | ✅ |
| Periodic run (supports setting Agent jobs to run periodically) | ❌ | ❌ | ✅ |
| MCP (supports configuring custom MCP for extensible Agent capabilities) | ✅ | ✅ | ✅ |
• Help document: https://www.alibabacloud.com/help/en/dms/data-agent-for-analytics/
• Data Agent: https://agent.dms.alibabacloud.com/
• Data Agent editions: https://www.alibabacloud.com/help/en/dms/data-agent-version-introduction#8205bd80afeko
ApsaraDB for ClickHouse Releases an Enterprise-Grade Agent Observability Solution
ApsaraDB - January 29, 2026
Alibaba Cloud Community - March 25, 2022
ApsaraDB - May 11, 2026
ApsaraDB - June 21, 2026
ApsaraDB - January 15, 2026
ApsaraDB - June 11, 2026
Token Plan
Build more, spend less. One plan, every modality.
Learn More
Alibaba Cloud Model Studio
A one-stop generative AI platform to build intelligent applications that understand your business, based on Qwen model series such as Qwen-Max and other popular models
Learn More
Hologres
A real-time data warehouse for serving and analytics which is compatible with PostgreSQL.
Learn More
Data Lake Analytics
A premium, serverless, and interactive analytics service
Learn MoreMore Posts by ApsaraDB
