- Home
- Data Federation Software
- MindsDB Query Engine
MindsDB Query Engine
Berkeley, California, United States
What it is
MindsDB Query Engine is a standalone open-source engine for querying connected data through SQL and running semantic search. The project says it connects to more than 200 data sources, including databases, warehouses, applications, and files; examples include PostgreSQL, MySQL, MongoDB, Snowflake, BigQuery, S3, and Slack. Queries run live through one SQL dialect without copying source data or using ETL pipelines. An HTTP SQL editor is included, and clients such as DBeaver and SQLAlchemy can connect through MySQL and PostgreSQL wire protocols. For unstructured content, knowledge bases combine an embedding model, an optional reranking model, and a vector store for indexing and retrieval. Hybrid search runs vector similarity and BM25 keyword matching in parallel, then merges results. Projects namespace work, views save cross-source transformations, and jobs schedule SQL runs. Deployment options documented by the project include Docker and PyPI installation. It supports Linux, web, API, and self-hosted use, and is described as a way to give AI agents live SQL access to data and semantic search. The project is free; MindsDB Core uses the Elastic License 2.0, and the repository notes that some directories have separate licenses.
For and against
For
- Queries connected sources live without copying data.
- Supports more than 200 data sources.
- Includes hybrid search for unstructured content.
- Docker and PyPI deployment are documented.
Against
- MindsDB Core uses the Elastic License 2.0.
- Some repository directories have separate licenses.
- Commercial support is directed to team contact.
The verdict
MindsDB Query Engine combines live cross-source SQL queries with semantic and hybrid search. Review its licensing terms and deployment approach to determine whether it fits your project.
Who it suits
MindsDB Query Engine may suit teams or developers who want to query connected data sources with SQL and add semantic search. Its documented deployment options and client protocols are relevant to users managing their own data environment.
Plans and prices
Open-source Query Engine
Standalone open-source project · MindsDB Core is licensed under Elastic License 2.0
Read 8 Oct 2026 · github.com
On the record
- Audience
- The project describes itself as a standalone open-source query engine that can give AI agents live, SQL-queryable access to data and semantic search.github.com · read 8 Oct 2026
- Deployment
- The project documents running the engine with Docker or installing it from PyPI.github.com · read 8 Oct 2026
- Federated queries
- It queries connected sources live through one SQL dialect without copying the data or using ETL pipelines.github.com · read 8 Oct 2026
- Hybrid search
- Hybrid search runs vector similarity and BM25 keyword matching in parallel and merges the results.github.com · read 8 Oct 2026
- Integrations
- The README names databases, warehouses, applications, and files as source categories and gives PostgreSQL, MySQL, MongoDB, Snowflake, BigQuery, S3, and Slack as examples.github.com · read 8 Oct 2026
- Interfaces
- The engine includes an HTTP SQL editor and supports MySQL and PostgreSQL wire protocols for clients such as DBeaver and SQLAlchemy.github.com · read 8 Oct 2026
- Knowledge bases
- Knowledge bases combine an embedding model, an optional reranking model, and a vector store to index and retrieve unstructured content.github.com · read 8 Oct 2026
- License
- MindsDB Core is licensed under Elastic License 2.0, and the repository notes that some directories have their own licenses.github.com · read 8 Oct 2026
- Maker history
- MindsDB's about page says Jorge Torres and Adam Carrigan released the first MindsDB version in Berkeley in 2018.mindshub.ai · read 8 Oct 2026
- Organization and automation
- Projects namespace work, views save cross-source transformations, and jobs schedule SQL to run on an interval.github.com · read 8 Oct 2026
- Purpose
- MindsDB Query Engine provides semantic search over data through SQL and connects to more than 200 data sources.github.com · read 8 Oct 2026
- Security and data handling
- The README says connected source data is queried in place and is never copied, and directs vulnerability reports to its security policy instead of public issues.github.com · read 8 Oct 2026
- Support
- The project points users to Discord for questions, GitHub Issues for bugs, and team contact for commercial support.github.com · read 8 Oct 2026
