---
id: dagster-mlflow
version: "0.29.18"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# dagster-mlflow — Package for mlflow Dagster framework components.

License: permissive · Maintenance: active · Downloads: 144.5K/mo

## What it is and what it does

dagster-mlflow is a library that bridges MLflow's experiment tracking and model registry with Dagster's asset-oriented orchestration framework. It allows you to instrument Dagster assets with MLflow tracking decorators so that model training runs, metrics, parameters, and artifacts are automatically logged to MLflow while remaining part of your Dagster lineage graph. This is useful when you want to track ML experiments and models alongside your data pipelines, keeping both the orchestration and the ML metadata in one place.

The package depends on dagster, mlflow, pandas, and protobuf. It is maintained as part of the main Dagster project and receives regular updates. The integration is designed to work within Dagster's declarative asset model, so you define your ML workflows as Dagster assets and use the integration to connect them to MLflow's tracking backend.

Use it for:

- Log model training metrics and parameters to MLflow while orchestrating the training job as a Dagster asset.
- Track model artifacts and versions in MLflow as outputs of Dagster asset runs.
- Build reproducible ML pipelines where experiment tracking is integrated into the asset lineage and observability.
- Manage model promotion and versioning through MLflow's registry while keeping orchestration logic in Dagster.
- Monitor ML experiment runs alongside data pipeline execution in a unified Dagster web UI.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Integrates MLflow experiment tracking and model management with Dagster's data orchestration, allowing you to track ML model training and artifacts as part of your Dagster asset pipelines.

Yes. The package is actively maintained, has no known vulnerabilities, and low install friction. It is worth installing if you are already using both Dagster and MLflow and want to unify experiment tracking with your data orchestration. If you use only one of these tools, it adds no value.

## Install

pip install dagster-mlflow
uv add dagster-mlflow
poetry add dagster-mlflow

## Installing dagster-mlflow

Before you install: Low friction install with a pure Python wheel. Active maintenance with a release on 2026-08-14 and 15996 GitHub stars indicate a well-maintained project. Depends on dagster, mlflow, pandas, and protobuf—all stable, widely-used packages.

License in practice: Apache-2.0 licensed under a permissive license, allowing use in commercial and proprietary projects with minimal restrictions beyond attribution.

Quickstart:

pip install dagster-mlflow

import dagster as dg
from dagster_mlflow import mlflow_tracking

@dg.asset
@mlflow_tracking
def my_model():
    return trained_model

Requires Python 3.10 or later (supports up to 3.14); dagster and mlflow must be installed as runtime dependencies.

Verify before relying:

- Specific MLflow version compatibility constraints beyond the runtime dependency declaration.
- Whether the integration supports all MLflow tracking APIs or a subset of common operations.
- Performance overhead of the integration when logging large model artifacts or many experiments.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 144.5K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

mlflow dagster integration, track ml experiments in dagster, mlflow asset logging, machine learning model orchestration, dagster ml pipeline, mlflow artifact management dagster, experiment tracking orchestration, ml-orchestration, experiment-tracking, data-pipeline

[View on SkillFed](https://skillfed.io/packages/dagster-mlflow) · [View on PyPI](https://pypi.org/project/dagster-mlflow/)