DVC
Git for data: version datasets and models like code.
About
DVC (Data Version Control) is Git for data and models: large files live in your cloud storage while tiny pointer files are versioned in Git — giving you reproducible pipelines, experiment comparison, and 'git checkout' for datasets. 15.9k stars, Apache-2.0, 1.6M PyPI downloads/month, VS Code extension included. Now under the treeverse org (moved from iterative; old URLs redirect). It adds real workflow complexity, so adopt it when dataset/model versioning is actually painful — for solo experiments, it's overhead.
What it can do
- **Data versioning**: Git-tracked pointers to cloud-stored data
- **Pipelines**: dvc.yaml stages with dependency tracking
- **Experiments**: track params, metrics, and plots in Git
- **Reproducibility**: `dvc repro` rebuilds exactly
- **VS Code extension**: visual experiment management
Setup
```bash
pip install dvc
dvc init
dvc remote add -d storage s3://mybucket/dvcstore
dvc add data/dataset.csv # then git add data/dataset.csv.dvc
```
Docs: dvc.org/doc.
Prerequisites
Python 3.9+; Git. Cloud storage (S3, GCS, Azure) or local remote.
Use cases
Works with
Trust score — 9/10
Scored 9/10 (verified tier, metrics-v1). Evidence: 15,911 GitHub stars (log-scaled vs catalog peers); blended with ~1,631,891 monthly package downloads; last push 0d ago; 8 releases/12mo; 221 open issues. Manual review: Git for data (version datasets/models, reproducible pipelines), 15.9k stars, 1.6M PyPI/mo; moved iterative→treeverse; adds workflow complexity for small teams. License: Apache-2.0.
★ 15,911 GitHub stars
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 8.2/10
- Maintenance · recency, releases, issue load
- 8.8/10
- Openness · license + docs
- 10.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · 15,911 GitHub stars (log-scaled vs catalog peers)
- · blended with ~1,631,891 monthly package downloads
- · last push 0d ago; 8 releases/12mo; 221 open issues
- · manual review: Git for data (version datasets/models, reproducible pipelines), 15.9k stars, 1.6M PyPI/mo; moved iterative→treeverse; adds workflow complexity for small teams
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.