ZenML

ZenML is an open-source MLOps framework that decouples pipeline logic from infrastructure so ML and LLM pipelines run portably from local to cloud, with 60+ integrations and a managed cloud from $49/month.

ZenML provides a standardized way to define ML and LLM pipeline steps, track experiments and artifacts, and deploy models across different infrastructure backends. Its central idea is portability: by decoupling pipeline logic from the underlying compute, the same code can run locally during development and then on Kubernetes, a cloud, or a managed ML platform in production without rewrites. The framework ships with 60+ integrations spanning PyTorch, scikit-learn, LangChain, LlamaIndex, Kubernetes, AWS, GCP Vertex AI, Kubeflow, and Apache Airflow, positioning it as an orchestration and glue layer across the MLOps and LLMOps stack. It handles experiment tracking, artifact lineage, and reproducibility so teams can move from notebook to production reliably. ZenML is Apache 2.0 licensed with an active GitHub community. Beyond the free open-source core, ZenML Cloud offers a managed control plane with a free tier and paid plans starting around $49/month, adding collaboration, managed servers, and enterprise controls for teams standardizing their pipelines.