---
id: databricks-feature-store
version: "0.17.0"
license: Databricks Proprietary License
license_treatment: unclear
maintenance: dormant
---

# databricks-feature-store — Databricks Feature Store Client

License: unclear · Maintenance: dormant · Downloads: 576.7K/mo

## What it is and what it does

databricks-feature-store is a deprecated client library for interacting with Databricks Feature Store. As of v0.17.0, the package is no longer maintained and has been superseded by databricks-feature-engineering, which provides the same functionality under a new package name. The library depends solely on databricks-feature-engineering at runtime, making it effectively a compatibility shim.

The package is licensed under a proprietary Databricks license that restricts use to connection with Databricks Platform Services under a separate agreement. It requires Python >=3.7 and has low install friction, but should not be used for new projects—existing code can migrate by installing databricks-feature-engineering, which maintains backward compatibility with the original import paths.

Use it for:

- Migrating legacy code from databricks-feature-store to databricks-feature-engineering without changing imports
- Accessing Databricks Feature Store within an existing Databricks workspace deployment
- Managing ML feature definitions and lineage on the Databricks platform

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

This package is deprecated as of v0.17.0; it provides a client for Databricks Feature Store but has been replaced by databricks-feature-engineering, which maintains backward compatibility with existing imports.

No. This package is explicitly deprecated as of v0.17.0 and dormant (948 days since last release). Install databricks-feature-engineering instead, which provides the same functionality and maintains backward compatibility with existing imports. The proprietary license also restricts use to Databricks Platform Services under a separate agreement.

## Install

pip install databricks-feature-store
uv add databricks-feature-store
poetry add databricks-feature-store

## Installing databricks-feature-store

Before you install: Low install friction, but the package is dormant (948 days since last release) and explicitly deprecated in favor of databricks-feature-engineering. New projects should use the replacement package instead.

License in practice: Licensed under Databricks Proprietary License (unclear treatment), which restricts use to connection with Databricks Platform Services under a separate agreement. Use is prohibited without an active agreement with Databricks.

Quickstart:

# This package is deprecated; use databricks-feature-engineering instead.
# pip install databricks-feature-engineering
# from databricks.feature_store import FeatureStoreClient
# client = FeatureStoreClient()

Requires active Databricks Platform Services agreement and connection to a Databricks workspace; proprietary license restricts use.

Verify before relying:

- Whether existing code using databricks-feature-store will continue to work without modification after switching to databricks-feature-engineering
- Whether the proprietary license permits use in open-source or non-commercial projects
- Specific Python version compatibility beyond the stated >=3.7 requirement

## Package facts

- License: Databricks Proprietary License (unclear)
- Python support: supports_current
- Install friction: low
- Maintenance: dormant
- Downloads: 576.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

databricks feature store client, feature engineering databricks, ml feature management, databricks ml features, feature store deprecated, deprecated, databricks-ecosystem

[View on SkillFed](https://skillfed.io/packages/databricks-feature-store) · [View on PyPI](https://pypi.org/project/databricks-feature-store/)