CLI Tool, MLOps Tool
15.5k 2026-03-27
treeverse/dvc
A command-line tool and VS Code extension for data versioning, ML experiment tracking, and reproducible machine learning pipelines.
Core Features
Version data and models, storing them in cloud storage while keeping metadata in Git.
Iterate fast with lightweight ML pipelines that only re-run impacted steps.
Track and compare ML experiments (data, code, parameters, metrics) locally without servers.
Share experiments and automatically reproduce anyone's results.
Quick Start
pip install dvcDetailed Introduction
DVC (Data Version Control) is an open-source command-line tool and VS Code extension designed to bring Git-like version control to data and machine learning models. It enables MLOps teams to manage large datasets and models, track experiments, and build reproducible pipelines without requiring dedicated servers. By integrating with existing Git workflows and cloud storage, DVC streamlines the development of machine learning projects, ensuring traceability, collaboration, and efficient iteration, akin to 'Git for data' and 'Makefiles for ML'.