# treeverse/dvc

🦉 Data Versioning and ML Experiments

Repository: https://github.com/treeverse/dvc
Canonical: https://ross.abutalabs.com/products/dvc
Homepage: https://dvc.org
Language: Python
License: Apache-2.0
License Family: permissive
Topics: data-science, machine-learning, reproducibility, data-version-control, developer-tools, ai, unstructured-data
Last push: 2026-08-24T18:51:06+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 77, longevity 100
- inputs: {"age_days": 3469, "days_push": 9, "days_rel": 155, "gap_med": 23.0, "n_releases_24m": 19}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 15843, forks 1323 (observed 2026-08-28T04:11:14.223163+00:00)

## What it is
DVC (Data Version Control) is an open-source command line tool and VS Code extension for versioning data and models, building lightweight ML pipelines, and tracking experiments in Git. It stores large data in cloud or remote storage while keeping version metadata in the Git repository, enabling reproducible machine learning projects.

## Use cases
- version large datasets and ML models alongside code in Git
- track and compare machine learning experiments locally
- build reproducible data science pipelines with dependency-aware stages
- share and reproduce teammates' experiments
- store data in S3, GCS, Azure, or other remote storage without committing to Git
- run CI/CD for machine learning projects

## When to choose
- you need Git-like version control for datasets and models too large for Git
- you want lightweight, local-first ML experiment tracking without servers
- you build reproducible ML pipelines with incremental re-runs
- you work in Python-based data science workflows and want minimal overhead

## When to avoid
- you need petabyte-scale, branch-and-merge data lake versioning (consider lakeFS)
- your project has no data or model artifacts to version
- you need a full hosted experiment tracking UI with team dashboards out of the box

## Facets
- artifact type: cli-tool
- maturity: active
- function: version-control, developer-tools, workflow-automation, data-science, machine-learning, configuration-management
- domain: data-science, machine-learning, developer-tools, version-control
- platform: windows, python, cli, cross-platform
- tags: data-version-control, mlops, experiment-tracking, pipelines, reproducibility, git, vs-code-extension, data-engineering, linux, macos

## Member repositories
- treeverse/dvc (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:14.223163+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:05:21.425197+00:00, confidence not recorded.
  - readme: https://github.com/treeverse/dvc (fetched 2026-08-28T04:11:14.223163+00:00, sha 34a820f2b2c3)
  - homepage: https://dvc.org (fetched 2026-08-29T08:03:23.288526+00:00, sha a49887c64378)
  - site_page: https://doc.dvc.org (fetched 2026-08-29T08:03:23.305926+00:00, sha 16c8e2cf4c9c)
  - site_page: https://dvc.org/blog/dvc-joins-lakefs-your-questions-answered (fetched 2026-08-29T08:03:23.308280+00:00, sha 8f9e94314905)
- Data as of 2026-08-30T08:39:29.467469+00:00.