MLOps & Experiment Tracking

MLflow

mlflow/mlflow

Popular

Open-source platform for experiment tracking, model registry and deployment, now with GenAI tracing and evals.

Weights & Biases

wandb/wandb

Popular

Hosted experiment tracking, sweeps, artifacts and Weave LLM tracing; owned by CoreWeave.

Optuna

optuna/optuna

Define-by-run hyperparameter optimization framework with pruning and a dashboard.

DVC

treeverse/dvc

Git-based data and model versioning with reproducible pipelines; now maintained by lakeFS (Treeverse).

Kubeflow

kubeflow/kubeflow

Kubernetes-native toolkit for ML pipelines, notebooks, training operators and model serving.

ZenML

zenml-io/zenml

Open-source MLOps framework for portable, reproducible ML and LLM pipelines across any infrastructure.

Metaflow

Netflix/metaflow

Netflix's human-friendly framework for building and scaling data science and ML workflows.

ClearML

clearml/clearml

Open-source suite for experiment management, orchestration, data management and model serving.

BentoML

bentoml/BentoML

Python framework for packaging models into production APIs and deploying them with autoscaling.

Amazon SageMaker

aws.amazon.com

Popular

AWS's fully managed service for building, training, tuning and deploying ML models.

Kedro

kedro-org/kedro

Kedro is a toolbox for production-ready data science. It uses software engineering best practices to help you create data engineering…

SkyPilot

skypilot-org/skypilot

The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build…