Get cell explanation with AI
Use the Explain cell feature to get an explanation of the underlying values that contribute to a specific cell value. Cell values can be a result of complex calculations, several consolidations, or rules. The Explain cell feature uses generative AI to summarize the cell intersection, underlying logic of the cell value, and source type.
To get a cell explanation, right-click a cell in a cube view. Select Explain cell [AI] to generate the explanation.

The generated explanation includes a table, with the following columns, and an AI-generated summary:
- Cell
- Cell identifies the original cell by its coordinates in the cube, by using a comma-separated list of member names. If there are members that contribute to the original value, you can drill down to view those members, their values, and source.
- Value
- Displays the value of the original cell or member that contributes to the value of the original cell. If the value is over 10,000, the value that is displayed is truncated to an integer, otherwise the value shows two decimal places. You can hover over the displayed value to view the actual raw value as it exists in the TM1 database. To view the actual values for all data, enable Display unrounded values.
- Source
- Identifies the source of the value: Consolidation, Input (a leaf cell that is not rule calculated), or rule. If the source is a rule, the entire rule is displayed. If the rule is too long to appear in its entirety within the confines of the column, click Expand row to reveal the full rule. You can also click the AI icon in each source row to generate an explanation for that member cell.
The explanation provides a descriptive summarization of the cell intersections along with underlying logic for the value held by the cell.
After you generate a cell explanation, you can add the summary to the related cell as a comment.
AI-generated summary
The Explain cell feature uses granite-3-8b-instruct from the Granite family of IBM foundation models to generate the summary.
- Sourced from quality data sets in domains such as finance (SEC Filings), law (Free Law), technology (Stack Exchange), science (arXiv, DeepMind Mathematics), literature (Project Gutenberg (PG-19)), and more.
- Compliant with rigorous IBM data clearance and governance standards.
- Scrubbed of hate, abuse, and profanity, data duplication, and blocklisted URLs, among other things.