Directory Data & AnalyticsBI & Data VisualizationCube.dev

Cube.dev

Open Source

About Cube.dev

Cube.dev offers an agentic analytics platform built on a semantic layer. The platform provides business intelligence for both people and AI agents. It uses one semantic model to keep answers governed and consistent. Cube gives AI agents governed, reliable business data through MCP and APIs.

Products

Analytics Chat: Builds queries against the semantic model from natural-language questions.
Workbooks: Combines SQL, a visual editor, and a Workbook Agent for drafting and iterating on queries.
Dashboards: Allows building dashboards conversationally with a Dashboard Agent.
Cube Cloud: Enterprise platform for internal BI and embedded analytics.
Analytics Chat API: Used to build a custom AI analytics experience that is agent-to-agent capable via MCP.

Used for

Building dashboards conversationally.
Connecting Cube to tools like Claude, ChatGPT, Cursor, or Slack via the MCP server.
Writing and iterating on queries using SQL and a visual editor.

Who it is for

Used by data teams.
Used by hobbyists completing small projects and showcasing work.
Used by professionals streamlining data workflows.

From Cube.dev's own website, read 2026-10-23. How we build listings.

How to evaluate BI & Data Visualization

This is how the CIOPages Research Team evaluates this category. It is not an assessment of Cube.dev. It comes from our Business Intelligence & Analytics buyer guide.

25%

Semantic Layer & Governed Metrics

How metric definitions are authored, versioned, and certified; whether the model is tool-locked or interoperable (LookML, dbt/Cube via SQL or MCP); row- and column-level security inheritance; lineage; and prevention of conflicting definitions across content

20%

Self-Service & Analytical Depth

Authoring experience for analysts vs. business users, drag-and-drop vs. spreadsheet vs. search paradigms, calculation expressiveness (DAX, LookML, set/associative logic), depth of ad-hoc exploration, and how far a non-technical user gets without IT

15%

AI & Natural-Language Analytics

Quality of NL-to-query and agentic features (Copilot, Pulse, Spotter, Qlik Answers, Gemini, Joule), whether answers are grounded in the governed semantic layer, explainability and citation of results, hallucination guardrails, and forecasting/insight automation

15%

Data-Platform Fit & Performance

Live query-down vs. in-memory extract behavior on your warehouse (Snowflake, Databricks, BigQuery, Fabric/OneLake), concurrency at your real user count, caching/acceleration, freshness, and whether one governed copy of data is preserved

10%

Embedding & Extensibility

White-label embedding, multi-tenant isolation and per-tenant row-level security, developer SDKs and REST/GraphQL APIs, infrastructure-as-code for content, and headless/API access for products you ship to customers

10%

Governance, Security & Administration

SSO/SAML/SCIM, granular RBAC, certified-content workflows and usage analytics, deployment options (multi-cloud, sovereign, on-prem), audit logging, and the operational burden of capacity, version, and content sprawl management

Total Cost & Ecosystem Alignment

Licensing unit (per-user vs. capacity vs. consumption) and how it scales to thousands of viewers, bundling with an incumbent cloud, migration and retraining cost, and lock-in to a single vendor’s data stack

Other BI & Data Visualization Vendors

CIOPages put this listing together from public sources. It’s information, not an endorsement. How we build listings. Work here? Claim this listing or .