lakeFS
Git-like data version control for data lakes and AI pipelines.
About
lakeFS is an open-source (Apache 2.0) data version control system that brings Git-like branching, committing, merging, and time travel to object storage data lakes on S3, GCS, and Azure Blob. It enables reproducible, auditable ML pipelines with zero-copy branching and atomic commits, plus hooks for data quality gates.
The project is very much alive: stable release 1.87.0 shipped September 22, 2026, Treeverse raised a $20 million growth round in July 2025 (bringing total funding to $43 million), and it acquired DVC in 2025. Customers include Arm, Bosch, Lockheed Martin, NASA, and Volvo.
The managed lakeFS Cloud Team plan starts at $499/month, so serious use is either self-hosted or paid. Self-hosting requires operating the service on your own infrastructure.
What it can do
- Zero-copy branching and merging on data lakes
- Atomic, immutable commits with time travel
- Pre-commit and pre-merge data quality hooks
- S3-compatible API and Spark/DuckDB integrations
Setup
Self-host via Docker/Kubernetes or sign up for lakeFS Cloud at lakefs.io; use the lakectl CLI.
Prerequisites
Object storage (S3, GCS, Azure Blob) and a data pipeline or ML workflow to version.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 72; Trending score 85/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.