Pricing

Ship ML pipelines and AI agents with confidence

Start open source and self-hosted. Upgrade to Pro for the managed control plane — on our SaaS or your own infrastructure.

Open Source

For individuals and small teams

Free

Unlimited executions · Unlimited projects


Includes

  • Pipeline & flow orchestration
  • Artifact management
  • Basic model registry
  • Community support
Recommended

Scale

SaaS

For teams running ML in production

$999 /month
Monthly executions
500 2,000 5,000

2,000 executions · 3 projects · 5 snapshots


Everything in Open Source, plus

  • Model Control Plane
  • Artifact Control Plane
  • Snapshots
  • Codespaces (remote IDE)

Enterprise

SaaS

For organizations at scale

Custom

Unlimited executions · Unlimited projects


Everything in Scale, plus

  • SSO (SAML / OIDC)
  • RBAC (custom roles)
  • Audit logs
  • Air-gapped deployment

Compare every plan

Open Source and Pro — with Scale and Enterprise feature tiers.

Plans Open SourceScaleEnterprise
Core platform
Pipeline & flow orchestration
Artifact management
Model registry (basic)
Pro control plane
Model Control Plane ↗
Artifact Control Plane ↗
Snapshots ↗
Codespaces (remote IDE)
Agent runtime
Durable execution for Python agents
Checkpoints, replay, wait/resume
Dashboard, API, schedules, webhooks
Distributed execution
Enterprise & governance
SSO (SAML / OIDC)
RBAC (custom roles) ↗
Advanced native scheduling
Audit logs
Air-gapped deployment
Support
Support level Community Priority Dedicated + SLA
Get Started Book a demo Talk to an engineer
What's included in Pro

Two products, one plan.

Switch SDKs without switching tools, billing, or governance. ZenML for reproducible ML. Kitaru for durable AI agents. Same control plane underneath.

ML pipelines

Pipelines & artifacts

Reproducible training, batch inference, evaluation. One DAG, versioned artifacts, every orchestrator.

  • Pipeline DAGs, artifact store, model registry Version every step, every dataset, every model.
  • Run on Kubernetes, Vertex, SageMaker, AzureML One pipeline, any orchestrator — no rewrites.
  • Reproducible by default, replayable on demand Re-run any historical pipeline with one command.
Agent runtime

Replay & checkpoints

Long-running Python agents with checkpoints, replay, wait/resume. Two decorators, no rewrites.

  • Durable execution for long-running Python agents Hours-long workflows survive restarts and failures.
  • Checkpoints, replay, wait/resume — two decorators Pause for human review, replay from any step.
  • Distributed scheduling, API and webhook triggers Fan out across workers, trigger from any source.

Same control plane. Same governance. Same bill.

Are you startup or academic?

Apply for a special price to access ZenML Pro features for early-stage companies building ML-powered products, universities, research institutions, and educational use cases.

Apply Now

No compliance headaches

Your VPC, your data

ZenML is a metadata layer on top of your existing infrastructure, meaning all data and compute stays on your side.

ZenML only has access to metadata; your data remains in your VPC
SOC 2 Type II BadgeISO/IEC 27001:2022 certification badge

ZenML is SOC2 and ISO 27001 Compliant

We Take Security Seriously

ZenML is SOC2 and ISO 27001 compliant, validating our adherence to industry-leading standards for data security, availability, and confidentiality in our ongoing commitment to protecting your ML workflows and data.

Support

Frequently asked questions

Everything you need to know about the product.

ZenML and Kitaru — same plan?
Yes. Pricing is unified across both workspaces. The ZenML workspace runs ML pipelines (typed step DAGs, training, batch inference). The Kitaru workspace runs durable AI agents (checkpoints, replay, wait/resume). You pick the workspace per project; Pro plans include both. Same $, same support tier, different SDKs and UI.
Can I self-host Kitaru like ZenML?
Yes. Kitaru is open source under Apache 2.0 — same model as ZenML. Self-host the server in your own VPC, point it at S3/GCS/Azure Blob, and run flows on Kubernetes, Vertex, SageMaker, or AzureML. Pro adds distributed execution, dashboard/API/scheduled/webhook triggers, SSO, audit, and a hosted control plane.
What happens if I exceed my plan’s limits?
We don’t abruptly cut you off. If you occasionally exceed your plan’s limits, we’ll notify you and suggest an upgrade if the pattern continues. For consistent overages, upgrading to the next tier ensures you get the best value.
Can I self-host ZenML?
Yes! ZenML is open source and can be self-hosted on your own infrastructure completely free — that's the Open Source plan. If you need the Pro control plane or support for a self-hosted deployment, the Enterprise plan covers that too. ZenML uses industry-standard encryption for all data in-transit and at rest. The data is stored on AWS regions in Europe, and a strict backup policy is maintained for all client data. ZenML only stores metadata — and no actual data is kept anywhere on our servers. Data and compute stays on the VPC of the customer.
How do Run Template Triggers work?
Run Template Triggers automate executions — ZenML pipeline runs and Kitaru flow executions — from events, schedules, webhooks, or the API. Triggers are a Pro feature: they're available on the Scale and Enterprise plans for more complex automation workflows in production.
What’s the difference between the managed plans and the open source version?
The Open Source plan gives you complete control but requires you to manage your own infrastructure. The Pro plans — Scale and Enterprise — provide a fully managed control plane with automatic updates, guaranteed uptime, and Pro-only features like run templates, the Model Control Plane, and RBAC. Enterprise adds governance and deployment options like SSO, audit logs, and air-gapped deployment on top.
What kind of support is included in each plan?
The Open Source plan includes access to our public Slack channel and community forums. The Scale plan adds priority support with faster response times. Enterprise includes a dedicated account manager, support with SLAs, and implementation assistance.

Trusted by 1,000s of members of top companies

Join the ZenML Community and start improving your MLOps

1,000,000

pipelines run in ZenML

100,000

pipelines run last month

21,000

stacks registered last 12 months

200,000

models trained last 12 months

"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."

Harold Giménez

Harold Giménez

SVP R&D at HashiCorp

HashiCorp
AirbusAXABundeswehrEnelJetBrainsBrevoCross Screen MediaADEOLeroy MerlinMann+HummelKoble

Start deploying reproducible AI workflows today

Enterprise-grade MLOps platform trusted by thousands of companies in production.