Malloy vs Sema
Malloy
malloydata.devSema
semalayer.comSide by side
| Figure | Malloy | Sema |
|---|---|---|
| Price from | — | $49/mo |
| Free plan | — | Yes |
| Free trial | — | Yes |
| Open source | — | — |
| Web | Yes | Yes |
| Windows | Yes | — |
| macOS | Yes | — |
| Linux | Yes | — |
| API | Yes | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 0 | 4 |
| Spec score · Semantic Layer Software | 7.6 (#18) | 8.7 (#4) |
| Support | Malloy directs users to its Slack community to share work, ask questions, and provide feedback, and to its GitHub project for bugs and ideas. | The Pro plan includes email support, Business includes priority support, and Enterprise includes a named customer success manager. |
Plans
In their own words
Malloy
Malloy is an open source language for describing data relationships and transformations. Users can define reusable calculations, table relationships and business logic in semantic models, then compile queries into SQL optimized for the connected database. The language supports nested datasets and automatically guards against common errors in sums, averages and counts. Supported databases include DuckDB, MotherDuck, BigQuery, Databricks, Snowflake, PostgreSQL, MySQL, Trino and Presto. DuckDB can read Parquet, CSV and JSON files without additional setup. Malloy offers a Visual Studio Code extension for modeling, querying, transformation and visualization, plus a browser notebook through github.dev that requires a GitHub sign-in. Its Python package loads models, compiles queries and runs statements with DuckDB and BigQuery. The CLI can run Malloy files, build persistent tables and automate transformations and pipelines. The project is free and open source, with web, editor extension, API and self-hosted options, among others. DuckDB connections can disable network access; sandboxed mode also limits file access to allowed directories.
Sema
Sema is a semantic layer for enterprise AI that connects databases and answers business questions in plain English, with governance controls. It maps tables, joins, metrics, and business terms into a typed graph; verified inferred joins can be reused in later answers. It can ask clarifying questions, show the SQL and graph nodes behind an answer, and select a chart. Users can save and schedule reports for email delivery and set metric alerts for thresholds or deviations. Available connectors include PostgreSQL, MySQL, MariaDB, SQLite, and SQLAlchemy URIs; ClickHouse and Amazon Redshift are in beta, while SQL Server, Snowflake, BigQuery, Databricks, and dbt metadata are on the roadmap. Sema can run in a customer VPC or its own environment; self-hosting uses Kubernetes and Helm with customer-provided Postgres and object storage. The free Sandbox costs nothing and limits use to one connector, one user, 100 monthly chat queries, and five saved reports, without scheduled delivery or alerts. Pro costs $49 a month per user, with a three-seat minimum and a 14-day trial.