OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Modeldb

Modeldb is an open-source system for versioning machine-learning models and tracking metadata and experiments across the model lifecycle. It records model ingredients such as code, data, configuration and environment, and provides tools for performance dashboards, report sharing and lifecycle tracking.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Modeldb — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Modeldb when an ML team needs a self-hosted record of model versions and experiment context, especially when results must be compared or models followed beyond development.

Who it is for

  • Machine-learning developers who need reproducible records of model versions and their ingredients.
  • Teams comparing experiment runs and sharing performance reports.
  • Organizations tracking models from development through deployment and live monitoring.

Documented uses

  • Record model versions alongside code, data, configuration and environment details.
  • Log experiment metadata, including metrics, artifacts, tags and user information.
  • Compare experiment performance in dashboards and share reports.
  • Track models across development, deployment and live monitoring.
Before you choose

Requirements & limits

The project is open source under the Apache-2.0 license. Its README documents self-hosted deployment and says to contact the project about a hosted version.

  • The documented quickstart uses Docker Compose and requires backend/config/config.yaml for the backend.
  • The Python client is distributed in the verta package; the project also lists a Scala client.
  • The README lists TensorFlow and PyTorch integrations, pluggable storage, and Docker and Kubernetes support.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Modeldb locally and log experiment runs

Start Modeldb with Docker Compose, install its Python client and send experiment metadata to the local service.

Before you start

  • Docker and Docker Compose
  • A cloned Modeldb repository or a manually created backend/config/config.yaml file
  • Python and the verta package

Steps

  1. Start the services with docker-compose -f docker-compose-all.yaml up.

  2. Install the client package with pip install verta.

  3. Connect a Python Client to http://localhost:3000, create a project and experiment, then log runs with hyperparameters and metrics.

  4. Open http://localhost:3000 to view the logged models in the Modeldb Web UI.

Expected outcome

A local Modeldb instance with experiment runs and metrics visible in its web interface.

Guide limitations

  • The backend requires backend/config/config.yaml; the README says to clone the repository before starting or create the file manually.
  • This quickstart documents a local Docker Compose setup and Python logging workflow.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
VertaAI/modeldb · Official homepage title · retained in its source language
How we review sourcesSuggest a correction