DVC

What is DVC?

DVC (Data Version Control) is an open-source tool that brings a Git-like workflow to datasets, models and machine learning experiments. Commands such as dvc init, dvc add and dvc push keep large files out of the Git repository while small tracking files stay in history, so a commit points to the exact data and model behind a result. Data scientists use it to reproduce training runs, compare experiments and share data with teammates through ordinary Git hosting such as GitHub, GitLab or Bitbucket. A VS Code extension keeps experiment work inside the editor, and DVC Studio is a web interface for visualizing and comparing experiments. In November 2025 lakeFS acquired the project from Iterative, which said DVC stays open source with no license change or paywalled features. The website points enterprise teams with petabyte-scale data lakes to lakeFS, while DVC itself targets individuals and small projects. There is no pricing page, because DVC is free to use.

Use cases

• Version datasets alongside code
• Track and compare ML experiments
• Reproduce model training runs
• Share data between team members
• Keep large models out of Git history
• Visualize metrics and parameters

Who it's for

• Data scientists
• ML engineers
• Researchers
• Small data teams

Why use DVC

• Data and model versions sit next to the code history, so any past experiment can be rebuilt from one commit.
• Git-style commands keep the learning curve short for developers who already use Git daily.
• Large files stay in your own cloud or remote storage, so repositories stay small and fast to clone.
• The project is open source and, per the lakeFS announcement, will not change license or put features behind a paywall.
• A VS Code extension and a web UI make experiment results easier to browse and compare.

DVC features

• Git-like commands such as dvc init, dvc add and dvc push
• Tracks datasets and models without storing them in Git
• Large files kept in cloud or other remote storage
• Experiment tracking tied to Git commits
• DVC extension for VS Code
• DVC Studio web UI for metrics, plots and team sharing
• Works with GitHub, GitLab and Bitbucket repositories
• DVCLive library for logging experiment metrics

DVC integrations

• Git
• GitHub
• GitLab
• Bitbucket
• VS Code
• MLflow
• Amazon SageMaker AI
• lakeFS

How to get started with DVC

1. Run dvc init inside a project that is already a Git repository.
2. Use dvc add to track a dataset or model, so DVC stores a small pointer file and Git ignores the large file.
3. Commit the pointer files to Git and run dvc push to send the data to remote storage.
4. Log metrics and parameters during training, then compare experiments in the editor or in DVC Studio.
5. Teammates pull the code and run dvc pull to get the matching data version and reproduce the result.

DVC limitations

• Aimed at small and medium data science projects; petabyte-scale data lakes are pointed to lakeFS
• Assumes the project already uses Git
• The VS Code extension is moving to the Treeverse publisher, so a reinstall will be needed

DVC pricing

DVC is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

DVC alternatives

The main DVC alternatives listed on Debuly are MLflow, Weights & Biases, Oxen.ai, Determined, OpenLIT and Pi Labs. The table compares pricing and platforms; order is A–Z.

1. De

   Determined

   Open-source deep learning platform for distributed training, hyperparameter tuning, GPU scheduling and experiment tracking.

   Developer platformsCoding
   Open sourceVisit Determined (official site)
2. 

   MLflow

   Open-source AI engineering platform for tracing, evaluating and deploying LLM agents, apps and classic ML models.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit MLflow (official site)
3. 

   OpenLIT

   Open-source, OpenTelemetry-native platform for tracing, evaluating and guarding AI agents, with cost and GPU monitoring

   Developer platformsCodingAutomation and agents
   Open sourceVisit OpenLIT (official site)
4. OA

   Oxen.ai

   One API for 200+ AI models plus versioned data repositories, fine-tuning and dedicated GPUs for building on your own data.

   Developer platformsImage generationVideoAudio and voice
   PaidVisit Oxen.ai (official site)
5. PL

   Pi Labs

   Build tunable scoring systems that judge, rank and improve AI outputs using natural-language and code-based criteria.

   Developer platforms
   Visit Pi Labs (official site)
6. 

   Weights & Biases

   Platform for tracking ML experiments and tracing, evaluating and monitoring LLM apps and agents with W&B Weave.

   Developer platformsCodingAutomation and agents
   FreemiumVisit Weights & Biases (official site)

DVC FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from dvc.org · fetched 23 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.