DagsHub

A unified repository for machine learning that layers DVC and MLflow onto a GitHub-like interface, making it the best choice for teams prioritizing open-source standards over proprietary silos.

Excellent for teams already committed to DVC and MLflow, weaker for those seeking a zero-configuration, 'no-code' experiment tracker.

Machine Learning Git Datasets Collaboration Data Analysis Open-Source Ai Development Mlops

Analysis based on product data, pricing structure, traffic signals, and public user sentiment.

Visit Website

Share this tool

Facebook X

LinkedIn Reddit

WhatsApp Telegram

Who Should Use DagsHub?

Typical users

Data scientists and ML engineers in research-heavy teams or open-source projects who need to version large datasets alongside code.

Maturity fit

scaling

Choose this if…

You want to avoid vendor lock-in by using open-source standards like DVC and MLflow
Your priority is seeing data, code, and experiments in a single unified UI
You need integrated data labeling without jumping between separate tools

Skip this if…

You find DVC's command-line complexity a barrier to your workflow
You require a fully managed, proprietary 'black box' solution with zero setup
Your organization forbids hosting data on third-party cloud platforms

About DagsHub

DagsHub is a collaboration platform designed specifically for the machine learning lifecycle. It acts as a specialized version of GitHub that handles not just code, but also massive datasets, model weights, and experiment logs by integrating open-source tools into a single interface.

Official profiles

X (Twitter) LinkedIn YouTube Discord

What it actually does

It provides hosted versions of DVC for data versioning, MLflow for experiment tracking, and Label Studio for data annotation. Users push code via Git and data via DVC to a central repository where the entire team can visualize changes, compare experiments, and review data labels.

What makes it different

Unlike Weights & Biases or Comet, which use proprietary formats, DagsHub is built entirely on open-source protocols. It doesn't try to replace Git or DVC; it provides the infrastructure to host them, ensuring you can migrate your entire project elsewhere if needed.

Git repository hosting for ML code DVC remote storage for large datasets and models Fully managed MLflow server for every project Integrated Label Studio for data annotation Dataset visualization and diffing Automated CI/CD pipelines for ML (DagsHub Actions) Interactive experiment comparison tables

Ratings across the web

Capterra 0 reviews

0.0/5

Ratings aggregated from independent review platforms.

Key Features

DVC Integration

Allows versioning of multi-terabyte datasets without bloating Git repositories.

Zero-Config MLflow

Provides a hosted MLflow tracking URI out of the box for every repository.

Data Engine

Enables querying, visualizing, and labeling data directly from the storage bucket.

Git-based Pull Requests for Data

Facilitates code-style reviews for dataset changes and model improvements.

Direct S3/GCS Mounting

Connects existing cloud storage buckets to the DagsHub UI without moving data.

DagsHub Actions

Triggers specialized ML workflows like model testing or data validation on every commit.

Integrated Label Studio

Lets teams annotate images, audio, and text within the project repository.

Pricing

Free

Unlimited public repositories
3 private repositories
10GB DagsHub storage
Integrated MLflow & Label Studio

Popular

Pro

$45 per month

Unlimited private repositories
100GB DagsHub storage
Priority support
Advanced data engine features

Team

$150 per user/month

SAML SSO
Role-based access control (RBAC)
Custom storage connectors
Dedicated support engineer

Enterprise

Custom annual

On-premises or VPC deployment
Unlimited storage options
Custom security compliance
White-glove onboarding

Pricing checked 4 months ago

Pricing guidance

Best plan for most users: The Pro plan is the sweet spot for individual researchers and small startups who need private repositories and more than 10GB of storage.

Free plan enough? Yes, for students, open-source contributors, or solo developers working on small hobby projects with fewer than 3 private repos.

Upgrade when:

When you exceed the 3 private repository limit
When your dataset grows beyond 10GB
When you need team-wide access controls and SAML

Watch out for:

Storage limits apply to DagsHub-hosted storage; external S3/GCS mounting may incur separate cloud costs
Free tier compute for DagsHub Actions is limited

Competitive with GitHub and Weights & Biases, though the per-user pricing for teams can become expensive for large organizations.

Pros & Cons

Strengths

No vendor lock-in
Because it uses Git, DVC, and MLflow, you own your data and metadata in standard formats, making it easy to leave the platform if requirements change.
Unified project view
Eliminates the 'context switching' tax by showing code, data versions, and experiment results in one tab rather than three different tools.
Generous free tier for open source
Provides significant value to the research community by offering unlimited public repositories with integrated tracking tools.
Streamlined data labeling
The Label Studio integration allows for a tight feedback loop between model error analysis and data re-labeling.

Weaknesses

DVC learning curve
DagsHub relies heavily on DVC; if your team isn't comfortable with DVC's CLI and concepts, the platform will feel difficult to use.

Affects: Beginner data science teams
UI performance with massive repos
Users report occasional lag when navigating repositories with extremely large file counts or complex directory structures.

Affects: Enterprise teams with legacy data structures
Fragmented documentation
While the core features are well-documented, troubleshooting the interaction between Git, DVC, and DagsHub can sometimes lead to 'dependency hell'.

Affects: DevOps and MLOps engineers

Real User Sentiment

Generally positive, with users praising the 'all-in-one' feel and the commitment to open-source standards, though some find the setup overhead of DVC annoying.

Users tend to like

The GitHub-like familiarity of the interface
Not having to set up an MLflow server manually
The ability to diff datasets and see what changed visually
Responsive community support on Discord

Users commonly complain about

DVC can be finicky and hard to teach to new team members
Occasional bugs in the Label Studio integration
The UI can feel cluttered compared to specialized experiment trackers

Recurring tradeoffs

You trade the 'instant setup' of W&B for the long-term flexibility of open-source DVC/MLflow.

Happiest users

Teams who already use DVC and want a centralized place to collaborate and visualize their work.

Often frustrated

Users who want a 'magic' solution for data versioning without learning the underlying command-line tools.

Use Cases

Collaborative Research

Multiple researchers sharing a single MLflow dashboard and DVC remote.

Open Source ML

Hosting public datasets and models with a community-driven contribution model.

Data Labeling Workflows

Moving from raw data to labeled datasets and trained models in one platform.

Model Auditing

Using the Git-style history to track exactly which data version produced a specific model weight.

Education

Teaching students the principles of MLOps using industry-standard open-source tools.

Frequently Asked Questions

Is DagsHub just a wrapper for DVC?

No, while it provides a remote for DVC, it also includes a hosted MLflow server, an integrated Label Studio instance, and a custom UI for diffing datasets and visualizing pipelines. It acts as the orchestration and visualization layer for these tools.

How does DagsHub compare to Weights & Biases (W&B)?

W&B is a proprietary, highly polished experiment tracker that is easier to set up but harder to leave. DagsHub is built on open standards (MLflow/DVC), offering more flexibility and data versioning capabilities at the cost of a slightly steeper learning curve.

Can I use DagsHub with my existing S3 bucket?

Yes, DagsHub allows you to connect your own S3, GCS, or Azure Blob Storage. This lets you use DagsHub's UI to explore and version data without actually moving your files from your existing cloud infrastructure.

Does the free plan include MLflow?

Yes, every DagsHub repository—including those on the free tier—comes with a fully managed MLflow remote tracking URI that you can use to log parameters, metrics, and artifacts.

What are the storage limits for data?

The Free plan includes 10GB of DagsHub-hosted storage. The Pro plan increases this to 100GB. If you connect your own external cloud storage (S3/GCS), those limits do not apply to DagsHub, though you will pay your cloud provider for that storage.

Can I host DagsHub on my own servers?

Yes, DagsHub offers an Enterprise version that can be deployed on-premises or within your own Virtual Private Cloud (VPC) for organizations with strict data privacy requirements.

Why trust this page?

This evaluation combines product positioning, pricing analysis, traffic and market signals, and public user sentiment into a single decision-support page. Content is generated editorially — not copied from the vendor's website.

Funding & Company

Founded

2019

Stage

Seed

Total Raised

$3.61M

Latest Round

Accelerator/Incubator (Jan 2025)

Notable Investors

Hatch (Singapore) Aviso Ventures Lemonade Stand (Estonia)

DagsHub has raised a total of $3.61M, primarily through Seed and accelerator rounds, with its most recent funding occurring in early 2025. This modest funding level suggests a focus on capital efficiency and product-led growth rather than aggressive, venture-backed scaling.

Full funding report medium confidence

Market Signals & Traffic

Estimated visits, global rank, geography, traffic sources, monthly visit trends, and organic search keywords (Similarweb)—on a dedicated page built for depth and search.

Estimated visits: 0
Global rank: —
Snapshot: May 2026
Traffic trend: Surging