NLQueries - 2026 Pricing, Features, Reviews & Alternatives

NLQueries is an open-source, self-hosted natural language to SQL (NL2SQL) engine built for developers, data engineers, and technical teams. It converts plain-English questions into SQL, validates that SQL against the live database schema before execution, runs queries in read-only mode, and returns structured answers. Beyond relational data, NLQueries ingests unstructured documents including PDFs, Word files, Excel spreadsheets, Notion pages, and Confluence spaces, delivering hybrid responses that span both databases and documents in a single query. Deployment is available via Docker, pip (Python 3.11 to 3.14), or from source, under a Business Source License 1.1 that converts to Apache 2.0 on June 4, 2030. The operator retains full control over data egress and LLM provider selection.

NLQueries connects to eight dedicated database connectors (PostgreSQL, MySQL, Snowflake, BigQuery, Redshift, SQL Server/Azure SQL, DuckDB, SQLite) plus a generic SQLAlchemy connector. A self-updating YAML knowledge base is built from real query history, schema introspection, column descriptions, and common query patterns, improving accuracy as teams use the system. SQL is validated via sqlglot before execution, and queries can be previewed without touching the database. A semantic cache serves repeated or similar questions (cosine similarity of 0.97 or higher) in under 50ms without an LLM round-trip. An MCP server with OIDC authentication and per-tool authorization allows querying databases directly from AI assistant clients such as Claude Desktop and Cursor. Additional capabilities include conversational follow-ups, dbt import, coverage reporting, and OpenTelemetry tracing.

The enterprise tier of NLQueries extends the core engine with a web UI featuring chat, agent dashboards, and a knowledge base editor; table-scoped agents with row-level filters; pinboards and scheduled chats; SSO (SAML), SCIM, MFA, and API keys; audit log UI with retention policies; LLM cost tracking and quotas; embedded and white-label analytics; REST API and webhooks; and priority support. The core engine is free for production and self-hosted use, making NLQueries accessible to organizations at any scale.