Wren AI overview and key facts (October 2026)

Wren AI is an open-source text-to-SQL engine that answers questions about company databases in plain language. A semantic layer describes tables and metrics so that the generated SQL follows business definitions. It suits data teams that want to run the engine themselves.

The open-source edition provides the context engine with modeling, the text-to-SQL pipeline, a CLI and an MCP server, without a user interface. Wren AI Cloud adds charts, dashboards and enterprise features. It connects to more than 20 data sources, including BigQuery, Snowflake, PostgreSQL, ClickHouse, Redshift and Databricks. The GitHub repository had 17,765 stars and a last push on September 25, 2026.

• Turns questions into SQLUsers ask a question in plain language and Wren AI generates SQL against the connected database, guided by the semantic model. Analysts answer ad hoc questions faster, and business users need less help.
• Keeps metrics consistentThe semantic layer defines tables, relationships and metrics once. Generated queries use these definitions, so answers match existing reports.
• Gives AI agents access to dataThe MCP server and CLI let AI agents query company data through the same governed context. Teams can add data access to their own agents without building a separate interface.

• Core, SDK and skills are licensed under Apache-2.0, with a published path-by-path license map.
• Works with any language model, so companies can use models they already have approved.

Wren AI is maintained by Canner, Inc., registered in Delaware, United States, according to its privacy policy, with teams in San Francisco and Taipei; the open-source project can be used worldwide without country ties.

Running the open-source edition means your team handles hosting, updates and model costs. Decide which language model receives table names and questions, since schema information leaves the database. The license file notes that future modules may use AGPL-3.0, so check the license of each component you deploy.

No reviews yet. Be the first to share how your team uses it.

Every review shows its verification level. We read each one before it goes live. How we check reviews