# mlrun/mlrun

MLRun is an open source MLOps platform for quickly building and managing continuous ML applications across their lifecycle. MLRun integrates into your development and CI/CD environment and automates the delivery of production data, ML pipelines, and online applications.

Repository: https://github.com/mlrun/mlrun
Canonical: https://ross.abutalabs.com/products/mlrun
Homepage: https://mlrun.org
Language: Python
License: Apache-2.0
License Family: permissive
Topics: mlops, python, data-science, machine-learning, data-engineering, experiment-tracking, model-serving, mlops-workflow, workflow, kubernetes
Last push: 2026-08-26T14:21:58+00:00

## Health v2 (maintenance only)
Score: 96/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 90, longevity 100
- inputs: {"age_days": 2558, "days_push": 7, "days_rel": 65, "gap_med": 3.0, "n_releases_24m": 59}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1692, forks 317 (observed 2026-08-28T04:05:22.703748+00:00)

## What it is
MLRun is an open-source MLOps and AI orchestration framework for building, training, deploying, and monitoring machine learning and generative AI applications across their lifecycle. It integrates with development and CI/CD environments, automating data pipelines, model serving, and experiment tracking on Kubernetes and multi-cloud infrastructure.

## Use cases
- orchestrate end-to-end ML training and deployment pipelines
- track experiments, data lineage, and model versions
- deploy real-time model serving endpoints with autoscaling
- automate LLM fine-tuning and serving workflows
- run distributed data processing and feature engineering jobs on Kubernetes
- add MLOps and monitoring to existing ML code with minimal changes
- manage batch and streaming inference pipelines in production

## When to choose
- you need an end-to-end MLOps platform covering training through serving and monitoring
- your team runs ML workloads on Kubernetes or multi-cloud/hybrid infrastructure
- you want experiment tracking, lineage, and model serving from one framework
- you need to productionize ML or gen AI pipelines with CI/CD automation

## When to avoid
- you only need lightweight experiment tracking without orchestration
- you run simple single-machine ML without Kubernetes or cloud infrastructure
- you prefer composing individual tools like MLflow, Kubeflow, and Seldon yourself
- your project cannot adopt a Python-centric SDK and platform services

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, workflow-automation, monitoring, deployment, etl, llm-training, llm-inference, scheduling
- domain: machine-learning, data-science, large-language-models
- platform: python, cloud, self-hosted
- tags: mlops, experiment-tracking, model-serving, orchestration, pipelines, genai, ci-cd, model-monitoring, feature-store, data-engineering, kubernetes, automation, docker

## Member repositories
- mlrun/mlrun (main) score 96

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:22.703748+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:38:14.338583+00:00, confidence not recorded.
  - readme: https://github.com/mlrun/mlrun (fetched 2026-08-28T04:05:22.703748+00:00, sha fdf2a606b1de)
  - homepage: https://mlrun.org (fetched 2026-08-29T11:14:18.648195+00:00, sha 7750683b8923)
- Data as of 2026-08-30T08:39:29.467469+00:00.