Data Studio#
Data Studio is the dataset workspace available from the Data Studio tab of every dataset repository. It brings together Dataset Viewer, SQL Console, and the Data Studio Agent for a selected configuration and split.
Explore before publishing#
Use Data Studio to answer practical questions about a dataset revision:
- inspect paginated rows, search and filter the indexed split, and open a durable row link in Dataset Viewer;
- use SQL Console for a read-only filter, aggregation, or quality check;
- share or embed SQL and download a result as CSV or Parquet;
- save a useful query for your own later review;
- ask the Data Studio Agent to explain the selected schema, preview values, data-quality questions, or suggest a SQL query.
The Agent requires sign-in and uses your MEGA Inference balance. Treat its answers and suggested SQL as a starting point: run the query and review the result before making a decision about the data.
Publish a data change#
Data Studio explores a published revision; dataset files remain versioned in the repository. To make a change durable, edit or upload the dataset files in the repository, commit the change to a branch or revision, validate it again in Data Studio, and tag the reviewed release. Update the dataset card whenever a schema, split, provenance, or quality characteristic changes.
This workflow preserves a clear connection between the result you inspected and the exact data that downstream users can download. See Uploading Datasets and Dataset Viewing for the corresponding workflows.

