"ZenML cuts out a lot of the infrastructure busywork so I can focus more on the tasks relevant to me. ZenML Pro helps us to keep an overview, as all components we need for MLOps, including documentation and metrics are available in one place!"
A managed control plane
for AI workflows and agents
ZenML Pro runs both workspaces under one paid product: ZenML for AI orchestration, Kitaru for agent insights and evals with our AI engineer on top. Single pane of glass, enterprise governance, no infrastructure setup.
Trusted by 1,000s of top companies to standardize their AI workflows
ZenML Pro · two sub-products, one plan
Managed control plane
Hosted infrastructure across both workspaces. You ship workflows; we run the plane.
- ZenML workspace
- Pipeline DAGs, artifact store, model registry, and a shared stack abstraction. Run on Kubernetes, Vertex, SageMaker, AzureML.
- Kitaru workspace
- Our AI engineer reads your traces next to your evals and shows you what they miss. Underneath, managed open-source Kitaru: import traces, get insights, turn findings into checks and replay a fix.
- Single pane of glass
- One control plane spans both workspaces. Pipeline runs and agent traces side by side in the dashboard.
Enterprise governance
What buyers need to ship in regulated environments, across both workspaces.
- SSO + RBAC + audit
- SAML / OIDC SSO and SCIM provisioning. Custom RBAC roles and org-wide audit logs. Same controls on ML and Agent workloads.
- Compliance
- SOC 2 Type II and ISO 27001. BYOK encryption. Multi-region and air-gapped deployment available.
- Support + SLA
- 24/7 dedicated support with a named solutions engineer and an SLA on response and resolution times.
"ZenML offers the capability to build end-to-end ML workflows that seamlessly integrate with various components of the ML stack, such as different providers, data stores, and orchestrators."
"ZenML allows you to quickly and responsibly go from POC to production ML systems while enabling reproducibility, flexibility, and above all, sanity."
"ZenML allows orchestrating ML pipelines independent of any infrastructure or tooling choices. ML teams can free their minds of tooling FOMO from the fast-moving MLOps space, with the simple and extensible ZenML interface."
A lot of our teams struggle to bring sanity to their model training processes. ZenML is built in a way that encourages good, maintainable pipelines.
ZenML allows you to keep your ML pipeline code cloud-agnostic, enabling faster future migrations to another technology stack.
Thanks to ZenML we’ve set up a pipeline where before we had only jupyter notebooks. It helped us tremendously with data and model versioning!
Get set up in less than a day
When you choose ZenML Pro, we don’t just hand over the keys. We make sure your infrastructure is ready, your pipelines are running, and your data scientists are onboarded.
01.
Guided Onboarding
Get personalized support to write your first AI workflows, ensuring a smooth start with ZenML.
02.
Seamless Infrastructure Setup
We handle the setup of your orchestration infrastructure, so you can focus on building exceptional models and agents. Bring your own cloud!
03.
Production Ready
Our experts will help your teams to connect your workflow and set up initial automations, streamlining your operations.
Your VPC, your data
ZenML is a metadata layer on top of your existing infrastructure, meaning all data and compute stays on your side.
SOC 2 Type II and ISO 27001
We Take Security Seriously
ZenML is SOC 2 Type II and ISO 27001 audited, validating our adherence to industry-leading standards for data security, availability, and confidentiality in our ongoing commitment to protecting your AI workflows and data.
ZenML Pro is Open Source and More
Built from the foundations the Open Source product, ZenML Pro is the best way to scale ZenML usage in your team.
01.
Managed control plane
ZenML Pro offers multi-tenant, fully managed deployments, with self-hosted and air-gapped options on Enterprise. Workspaces come in two types, ZenML for AI orchestration and Kitaru for agent evals, under the same organization, roles, and billing, with dev, staging, and production servers separated.
02.
Roles and Permissions
ZenML Pro tenants have built-in roles and permissions, as an extension to the open-source product. We connect ZenML with your OIDC provider and offer SSO.
03.
Control and configurability
ZenML Pro control plane allows you to run ZenML pipelines directly from the server, and features enhanced configurability for your pipeline builds.
04.
Enhanced observability
ZenML Pro tenants have an enhanced dashboard with more features including a model control plane to view all your models, and the ability to trigger pipelines, do CI/CD and lots more.
Frequently asked questions
Does ZenML Pro cover the Kitaru agent workspace too?
Yes. ZenML Pro is the paid plane over both sub-products: ZenML for AI orchestration and open-source Kitaru for agent insights and evals, with our AI engineer reading your traces next to your evals. Same control plane, same SSO/RBAC/audit, same compliance posture (SOC 2 Type II, ISO 27001, BYOK, multi-region). Per-project: pick the ZenML workspace, the Kitaru workspace, or both.
What about data security and compliance?
ZenML is designed architecturally in a way that no data flows through the actual ZenML Pro side, but remains inside your own cloud infrastructure behind your VPC. Only metadata is transferred to our servers (names of pipelines, models, statuses, metrics, etc), which we store encrypted at transit and at rest.
We are SOC 2 Type II and ISO 27001 audited (see our trust center), validating our adherence to industry-leading standards for data security, availability, and confidentiality in our ongoing commitment to protecting your AI workflows and data.
Please refer to our terms of service and privacy policy for more details.
We are SOC 2 Type II and ISO 27001 audited (see our trust center), validating our adherence to industry-leading standards for data security, availability, and confidentiality in our ongoing commitment to protecting your AI workflows and data.
Please refer to our terms of service and privacy policy for more details.
How does ZenML Pro interface with my internal services?
As no data enters ZenML servers, clients can run most actions with local authentication that interfaces with any internal services and the ZenML server is really only a metadata storage for your pipelines, models, runs etc. ZenML also offers a secure way of communicating with external services using service connectors, which are a really cool way of generating short lived tokens to talk to external services like Kubernetes, Docker registries, Blob storage etc. Read more here.
What are the various deployment scenarios for ZenML Pro?
ZenML Pro deploys the OSS server as a service, which centrally tracks everything related to pipelines, stacks, models, artifacts, etc. This page goes through various deployment scenarios for this deployment.
I am an open source user. Can I transition to the Pro version? What about my data!
Yes! Once you request a demo, we offer a migration service that helps transition your legacy database into your Pro account.
What is the difference between the open-source and Pro product?
ZenML is and always will be open-source at its heart. The core framework is freely available on GitHub and you can run and manage it in-house without using the Pro product. On the other hand, the Pro product offers one of the best experiences to use ZenML, and includes a managed version of the OSS product, including some Pro-only features that create the best collaborative experience for many companies that are scaling their ML efforts.
You can see a more detailed comparison here.
You can see a more detailed comparison here.
Unify Your ML
and LLM Workflows
Open-source foundation, no vendor lock-in · Works with any infrastructure · Upgrade to managed Pro features