Cube

Overview

Cube is an analytics platform for teams that want dashboards, embedded products, and AI agents to work from the same governed business model. It suits data teams building shared metrics for internal or customer-facing analytics; its request quotas and short history retention make plan choice important.

Cube Cloud pairs a managed production platform with the open-source Cube Core project. Core processes API requests using pre-aggregated data in Cube Store or raw data in connected databases, giving teams a path to serve analytics through an API rather than only through a dashboard.

Cube was founded in 2019 by Cube Dev, Inc.; the company's co-founders are Artyom Keydunov and Pavel Tiunov.

Key features

• Governed semantic layer: Shared business definitions keep metrics consistent across dashboards, embedded applications, and AI agents. This is useful where teams need one model to serve multiple analytics surfaces, though it depends on investing in that model rather than defining terms independently in each tool.
• Free analytics tools: The Free plan includes a semantic layer IDE, workbooks, dashboard publishing, semantic model and workbook agents, and analytics chat. That is a substantial starting toolkit, but the 1,000 daily request cap and five-workbook limit constrain sustained or broader use.
• Data and transformation integrations: Sources include PostgreSQL, Snowflake, ClickHouse, Databricks, BigQuery, MySQL, Oracle, Redshift, MongoDB, and Trino. Transformation and orchestration integrations include dbt, Dagster, Prefect, Airflow, and Coalesce.io, making Cube relevant to teams already using those systems.
• API processing and caching: Cube Core can process requests against pre-aggregations in Cube Store or raw data in connected databases; Cube Store stores those pre-aggregations. This supports different query paths, but the Free plan's request allowance remains a practical ceiling.
• Security and deployment: Cube states that data is encrypted at rest, API access uses token-based security, and traffic is encrypted over SSL/TLS 1.2. Cube Cloud states SOC 2 Type II, HIPAA, and GDPR compliance. Deployment choices include managed multi-tenant AWS or GCP, a VPC on AWS, GCP, or Azure, and BYOC on a customer's cloud, which gives organizations options ranging from managed service to customer-cloud hosting.

Pricing

Cube uses a freemium model, with paid plans from $40/mo. The tiers differ not just in request volume but also in retention and support, so teams should weigh those limits against their analytics workload.

• Free — free, free forever: Up to 5 workbooks, 1,000 daily requests, 1 day of query history retention, and community support. It is a sensible way to build and explore analytics, but the request and workbook caps make it a constrained choice for wider production use.
• Starter — $40 a month, per developer / month: 10K daily requests shared or 50K dedicated, 1 day of query history retention, and limited chat and email support. It suits a small team that needs more request capacity, but shared capacity may be less predictable across users and retention remains at one day.
• Premium — $80 a month, per developer / month: Unlimited queries, 7 days of query history retention, separately priced Explorer and Viewer roles, and chat and email support. It fits teams whose request volume or history needs outgrow Starter; the additional separately priced roles matter when many people need to explore or view analytics.
• Enterprise — custom pricing: Custom query history retention, a dedicated single-tenant installation, and a dedicated CSM. This is the fit for organizations that need those deployment and support terms rather than the standard tiers.

Platforms

Cube supports API, Linux, self-hosted, and web environments. Cube Cloud's AWS and GCP managed service, VPC deployments on AWS, GCP, or Azure, and BYOC option make it more flexible than a cloud-only service, especially for teams with deployment constraints.

Who it's for

Cube is best suited to data teams that need governed metrics to power internal BI and embedded analytics from one semantic model, particularly when they want APIs and options for cloud or customer-cloud deployment. It is less compelling for a small or low-volume use case that can operate within simpler request needs, or for teams unwilling to manage shared definitions and choose a tier around retention and quotas.

Pros and cons

• Pro: One governed semantic layer can serve dashboards, embedded analytics, and AI agents, reducing the need to maintain separate definitions for each surface.
• Pro: The Free plan includes authoring, dashboard publishing, agents, and chat, offering a broad set of capabilities before a paid commitment.
• Pro: Managed, VPC, and BYOC deployment options address different cloud and hosting requirements.
• Con: Free is capped at 1,000 daily requests and five workbooks, with only one day of query history, limiting its headroom for growing use.
• Con: Starter keeps one-day retention, while Premium prices Explorer and Viewer roles separately, making the full cost and fit depend on usage and audience.

Alternatives

• Semantic Layer Software is a useful category comparison for teams evaluating the broader semantic-layer field.
• Data Transformation Tools is a relevant category for readers whose priority is transformation and orchestration rather than analytics delivery.
• Tableau is another freemium option with a Free plan; choose it when its creator and explorer licensing structure is a better fit for the team.
• Starburst Enterprise Context Layer offers a free tier with up to three clusters and standard ad hoc query execution; consider it when that cluster-based starting point better matches your needs.
• dbt Semantic Layer has a free Developer plan with one seat, 3,000 successful models per month, and one project; consider it when those limits suit the intended use.
• Sema has a Sandbox plan limited to one connector, one user, 100 chat queries a month, and five saved reports; it is an option when that bounded sandbox is enough.
• Strata offers a Free plan for one developer and 25 users, with two data sources per project; consider it when those allowances fit.
• AtScale offers Standard semantic models and BI-tool connectivity, making it an option for teams focused on those capabilities.
• Kyvos is a paid alternative with a free trial and cloud marketplace pricing at $0.41 a month, billed at $0.41 per core hour for the time used.
• Definite is another freemium option with a Free plan and web platform support.

Verdict

Choose Cube if your team needs one governed semantic model to power BI, embedded analytics, and AI-facing workflows, with deployment options beyond a single managed cloud setup. Look elsewhere if you need generous history retention on an entry tier or want to avoid request caps and separately priced viewer roles.