Install unitycatalog with Homebrew, chocolatey, winget

Project links

• Homepage
• Repository
• Upstream docs
• Package manager page

Commands and aliases

• start-uc-server
• uc

Project history

The official documentation describes Unity Catalog as an open, multimodal catalog with an OpenAPI specification and Apache-2.0 implementation. It is designed to support multiple table formats, compute engines, and asset types, including tabular data, files, functions, and AI models.

The project launched publicly in July 2024. By the 0.2 release, the official blog highlighted external authentication, MLflow model support, Apache Spark and Delta Lake integration, and a new UI. The docs also note hosting under the LF AI & Data Foundation as a sandbox project.

Adoption history

Unity Catalog's adoption story is tied to the lakehouse and AI-governance ecosystem rather than to a single CLI audience. The upstream README lists ecosystem support from cloud providers, query engines, data tooling vendors, AI tooling projects, and governance vendors.

For local package users, adoption starts with the quickstart workflow: run the UC server locally, then use the uc CLI to inspect and operate on catalog content. That makes the Homebrew package valuable as a reproducible local test harness for an otherwise infrastructure-oriented system.

How it is used

The official quickstart shows a two-terminal workflow: start the Unity Catalog server, then use the CLI to list catalogs and schemas, operate on Delta tables, and exercise sample assets. Integration docs cover Spark, Delta Lake, MLflow model registration, Spice, LiteLLM, and AI function tooling.

Configuration is server-centered. Official auth docs point users at etc/conf/server.properties for identity-provider settings and describe local authorization data, while the quickstart and server docs keep the CLI/server pairing visible.

Why package nerds care

Unity Catalog is package-nerd significant because it gives data engineers a local, open source implementation of a catalog API that otherwise lives in managed data-platform environments. A package manager can turn a governance server plus CLI into a fast local sandbox.

It also reflects a newer packaging pattern: CLI tools for AI and data infrastructure increasingly bundle local servers, OpenAPI-compatible services, sample data, auth configuration, and integrations with heavyweight runtimes such as Spark and MLflow.

Timeline

• 2024: Unity Catalog launched publicly as an open source project.
• 2024: Version 0.2 added external authentication, MLflow model support, Spark and Delta Lake integration, and a UI.
• 2025: Documentation described Unity Catalog as an LF AI & Data Foundation sandbox project with open APIs and multimodal data and AI asset support.

Related projects

• Related technologies in the official docs and README include Apache Spark, Delta Lake, Apache Iceberg REST catalog compatibility, Apache Hive metastore API compatibility, MLflow, Spice OSS, LiteLLM, and table formats such as Delta, Iceberg, Hudi, Parquet, JSON, and CSV.
• The project also has a Rust implementation effort, unitycatalog-rs, which points to interest in native catalog implementations beyond the primary repository.

Sources

• https://docs.unitycatalog.io/
• https://docs.unitycatalog.io/quickstart/
• https://docs.unitycatalog.io/server/auth/
• https://github.com/unitycatalog/unitycatalog
• https://unitycatalog.io/blogs/unity-catalog-0-2-introduces-models-mlflow-and-spark-integration-and-support-for-external-identity-providers/
• source_facts.package-manager