All Products
Search
Document Center

DataWorks:DataWorks ChatBI overview

Last Updated:Mar 26, 2026

DataWorks ChatBI uses Natural Language Processing (NLP) and a Large Language Model (LLM) to automate the full data analysis pipeline — from understanding your question and identifying the right Data Table to generating Code, building a Chart, and surfacing key findings — all through a conversational interface.

Important

DataWorks ChatBI is in Public Preview and free to use. Starting April 1, 2026, it will transition to a paid model. Pricing details will be published on the official website before the commercial launch.

Prerequisites

Before you begin, make sure you have:

Requirement Details
Serverless Resource Group A Serverless Resource Group in the region where your DataWorks resources are located

How it works

image

Ask a question in the session window. ChatBI then runs the following steps automatically:

  1. Identifies the relevant Data Table in your Dataset

  2. Generates and runs query Code against the data

  3. Creates a Chart to visualize the results

  4. Extracts key findings and compiles a report combining text and visuals

ChatBI currently uses the Qwen3-235B-A22B LLM.

Access points

Log in to Alibaba Cloud and open the DataWorks ChatBI page. Select the access point that matches the region of your DataWorks resources, such as your resource group and Dataset.

China (Hangzhou)

China (Shanghai)

China (Shenzhen)

China (Hong Kong)

China (Chengdu)

China (Beijing)

China (Zhangjiakou)

Indonesia (Jakarta)

Get started

Follow these steps to run your first analysis.

Step 1: Create a Dataset

Create a Dataset to define the data scope for your analysis. A Dataset can be a Data Table from a target data source or a local file you upload.

Step 2: Start a ChatBI session

Start a ChatBI session and type your question. ChatBI identifies the target Data Table, queries the data, generates and runs Code, builds a Chart, and extracts key findings — all automatically.

Step 3: Improve results with a Knowledge Base (optional)

If session results are inaccurate or incomplete, create a Knowledge Base to give the LLM more context about your data and business. With a Knowledge Base, you can:

  • Add question templates to guide how ChatBI interprets common queries

  • Define Glossary Term entries so domain-specific language maps to the correct data

  • Encode business logic so the LLM generates Code that reflects your actual rules

Step 4: Share reports (optional)

Share reports generated by ChatBI as an HTML link or export them as a PDF for distribution to team members and stakeholders.

Billing

DataWorks ChatBI is in Public Preview and free to use. Starting April 1, 2026, it will transition to a commercial billing model. Detailed pricing plans will be announced on the official website before the commercial launch.