In Metabase
You want a self-hostable open-source BI tool over an existing warehouse.
Comparison
nlqdb vs Metabase
Pick Metabase if you want an open-source BI tool to build dashboards and let analysts ask charts in chat over your existing warehouse. Pick nlqdb if you're building a product or agent that needs English-to-SQL over a database it provisions — embeddable, API-first, every write diff-previewed.
The same goal, two ways.
> top 10 products by revenue this quarter, with month-over-month growth
In your HTML
<nlq-data goal="top 10 products by revenue this quarter, with month-over-month growth"></nlq-data>The kind of question an analyst would build a Metabase dashboard for — nlqdb answers the English goal as SQL over a database it owns, returned as rows your product embeds, not a dashboard you log in to.
What's different
Four dimensions that decide it
| Dimension | nlqdb | Metabase | Note |
|---|---|---|---|
| Owns the database (provisions + migrates) | Metabase connects to an existing warehouse to read it; it doesn't provision or own a database your app writes to. | ||
| Natural-language data questions | Metabot (paid add-on) answers in English and writes SQL; the OSS edition is basic single-shot SQL generation. nlqdb compiles SQL against a Postgres it owns on every plan. | ||
| Embeddable in your product (HTML element / SDK / API) | Metabase embeds finished dashboards/charts (iframe/SDK); nlqdb embeds an answer primitive — `<nlq-data>` takes an English goal and returns rows you render in your own layout. | ||
| MCP server (agent-callable) | nlqdb's `nlqdb_query` materialises Postgres on first reference for a Claude / Cursor agent; Metabot answers inside Metabase (and Slack), not as an agent-callable database. |
Show 5 more rows
| Dimension | nlqdb | Metabase | Note |
|---|---|---|---|
| Dashboards, charts + scheduled reports | Dashboards and visualizations are Metabase's home turf; nlqdb returns typed result rows you render in your own UI. | ||
| Open source / self-hostable | Metabase ships an AGPL open-source edition you self-host; nlqdb is source-available on FSL 1.1-ALv2, auto-converting to Apache 2.0 after two years. | ||
| Auto-migration via NL ('add a column for tags') | |||
| Destructive-op diff preview before apply | Metabase reads and visualizes; it doesn't manage your schema. nlqdb previews writes and DDL before applying. | ||
| Read across many connected sources | Metabase connects to many warehouses for read-only BI; nlqdb provisions and queries its own Postgres rather than federating external sources. |
shipped · partial · not shipped
When to choose nlqdb
- You're embedding data features into a product or agent, not building dashboards.
- You want one HTML element (`<nlq-data>`) or an API, not a BI app to log into.
- An AI agent must query — and provision — its own database, callable over MCP.
- Writes and schema changes should be diff-previewed before they apply.
When to choose Metabase
- You want a self-hostable open-source BI tool over an existing warehouse.
- Your team builds and shares dashboards, charts, and scheduled reports.
- Analysts want a familiar query builder with an AI assistant alongside.
- You need read-only analytics across many connected data sources.
Questions buyers ask
- Can I use Metabase and nlqdb together?
- Yes — they serve different jobs. Metabase is where analysts build dashboards and explore an existing warehouse; nlqdb is the database your product or agent queries in plain English at runtime. Use Metabase for BI reporting, nlqdb for the embedded data layer your app ships on.
- How is Metabot different from nlqdb's natural-language querying?
- Metabot is an AI assistant inside the Metabase BI app — it answers questions, builds charts, and writes SQL for analysts working in Metabase. nlqdb is a backend: it compiles English to SQL over a Postgres it provisions and owns, and exposes that through an HTML element, SDK, API, and MCP server you build on.
- Does nlqdb build dashboards like Metabase?
- No. nlqdb returns typed result rows from SQL it compiles; it doesn't build dashboards, charts, or scheduled reports. If shareable BI dashboards over your warehouse are the goal, Metabase is the right shape; nlqdb's contract is the data, which you render in your own UI.
- Is Metabot free in the open-source Metabase?
- Only partly. The open-source Metabase edition includes basic single-shot SQL generation, but the full Metabot — natural-language questions, chart building, error fixing, chart analysis — requires a paid Cloud plan plus a $100/month add-on (500 requests), or an Enterprise contract. nlqdb's English-to-SQL works on every plan, including the free LLM chain.
- Can an AI agent call Metabase the way it calls nlqdb?
- Not as a database. Metabot answers inside Metabase and Slack for human analysts; it isn't an agent-callable data layer. nlqdb ships an MCP server, so a Claude or Cursor agent can provision a Postgres and query it in English — `nlqdb_query` materialises the database on first reference.
Try nlqdb in 30 seconds
No sign-in. The anonymous database lasts 72 hours; adopt it with one click if you keep it.
Start with a goal →