ZenML
zenml.io
- Free plan
- Windows
- macOS
- Linux
- Web
- API
- Self-hosted
What ZenML is
ZenML orchestrates AI and machine-learning pipelines and agents written in Python. Pipelines can run on Kubernetes, Vertex AI, SageMaker, AzureML, or a laptop without rewriting code. ZenML versions step results, including models and evaluation datasets, and supports rerunning from any step. It can reuse cached step results when code and inputs have not changed. The integrations directory lists 66 tools, including LangGraph, OpenAI Agents SDK, PyTorch, MLflow, AWS, and Google Cloud Vertex AI Pipelines. Deployment options include local, SaaS, and self-hosted setups, with Docker and Kubernetes Helm deployment methods. ZenML says it stores metadata while data, artifacts, and compute stay in the customer's infrastructure or VPC. Its Open Source plan is free and includes unlimited executions and projects for self-hosted use. Scale costs $999 a month and includes 2,000 executions, three projects, and five snapshots. Enterprise pricing is custom and adds SSO, custom-role RBAC, audit logs, and air-gapped deployment. ZenML says it is SOC 2 Type II and ISO 27001 certified.
Who it suits
The free Open Source plan is described for individuals and small teams. Scale is aimed at teams running ML in production, while Enterprise is for organizations operating at scale.
Good to know, and what to watch
- Pipelines can run across several listed infrastructures.
- Step results can be versioned and rerun.
- Unchanged steps can reuse cached results.
- Self-hosted free plan includes unlimited executions and projects.
- Data and compute can remain in the customer's infrastructure.
- Scale is $999 a month.
- Scale includes 2,000 executions per workspace.
- Enterprise pricing is custom and not listed.
- Managed compute is not offered.
Plans
Every plan and priceEnterprise
Open Source
Scale
Free plan
The verdict
ZenML combines pipeline orchestration, versioned artifacts, caching, and multiple deployment options for Python-based AI and ML work. Teams should weigh Scale's execution and project allowances against their needs, or consider self-hosting on the free plan.
What the maker publishes
- Artifact versioning
- ZenML versions step results, including models and evaluation datasets, and supports re-execution from any step. source
- Caching
- ZenML caches step results and reuses them when code and inputs have not changed. source
- Data handling
- ZenML says it stores metadata while data, artifacts, and compute stay in the customer's infrastructure or VPC. source
- Deployment
- ZenML supports local, SaaS, and self-hosted deployment, including Docker and Kubernetes Helm deployment methods. source
- Enterprise controls
- The Enterprise plan adds SSO, custom-role RBAC, audit logs, and air-gapped deployment. source
- Headquarters
- Munich, Germany source
- Infrastructure
- Pipelines can run on Kubernetes, Vertex AI, SageMaker, AzureML, or a laptop without rewriting the code. source
- Integrations
- The integrations directory lists 66 tools, including LangGraph, OpenAI Agents SDK, PyTorch, MLflow, AWS, and Google Cloud Vertex AI Pipelines. source
- Overage
- ZenML says it does not cut off service for occasional monthly execution overages and may contact customers if overages become a pattern. source
- Purpose
- ZenML orchestrates AI and machine-learning pipelines and agents written in Python. source
- Security
- ZenML states that it is SOC 2 Type II and ISO 27001 certified. source
- Support
- The Open Source plan includes community support, Scale includes priority support, and Enterprise includes dedicated support with an SLA. source
- Target users
- The Open Source plan is described for individuals and small teams, Scale for teams running ML in production, and Enterprise for organizations at scale. source
Checked against the maker's pages on 30 Sept 2026.