Home / Directory / AI SaaS tooling / MLOps & experiment tracking / ZenML

ZenML

ZenML is a Munich company, now called ZenML Labs, that makes the open-source ZenML framework for ML and AI pipelines, the Kitaru agent testing tool, and ZenML Pro, a hosted platform with AI engineers that watch agents and pipelines.

8.0/10
Overall Score
Highly recommended

ZenML gives teams a free, open-source pipeline framework that runs on the infrastructure they already have, and its $39 Pro Starter plan is one of the cheapest ways to add hosting, schedules, and alerts.

Best for

Small and midsize ML teams that want open-source pipelines portable across clouds and orchestrators, and teams that now need to test and improve AI agents with the same tooling.

Not ideal for

Teams that want a turnkey managed compute platform with GPUs included, or buyers who need a long track record of large enterprise deployments.

Verdict

ZenML started as an open-source framework that lets data scientists write ML pipelines once and run them on Kubeflow, Airflow, SageMaker, Vertex AI, or local machines, and in 2026 it added Kitaru and AI engineers that watch production agents, find failures, and test fixes. It is highly recommended for teams that want portable open-source pipelines at a low published price, and it is a lighter fit for buyers who want compute and GPUs bundled into the contract.

Score Breakdown

These scores cover how well a team can log and compare experiments, how well it versions data, models, and other artifacts, how easily it runs work across clouds and clusters, and how well it fits enterprise security and audit needs, and then how the company does on innovation, communication, pricing, and contract terms.

Buyer outcomes

Experiment tracking
8.0
Artifact & registry
7.9
Orchestration / hybrid compute
8.4
Enterprise governance
7.8

Company & commercial

Innovation & product leadership
8.4
Project management & communication
8.0
Pricing
8.3
Contract fairness
7.2

Who Uses ZenML, Where It Fits on Price, and Its Competitors

ZenML is used by ML engineers and data scientists at software companies, retailers, and marketing platforms. Its customers include Brevo, ADEO (Leroy Merlin), Cross Screen Media, and Zuiver. The company is a remote team with an office in Munich, its founders have built together for ten years, and it raised a $6.4 million seed round led by Point Nine and Crane Venture Partners.

The open-source framework is free, Pro Starter costs $39 a month for one AI agent and one ML project, and Enterprise is quoted, which puts ZenML at the low end of this field on cost. It is most often compared with Union.ai and Kubeflow for pipeline orchestration, with MLflow and Weights & Biases for experiment tracking, and with ClearML for an open-source MLOps suite.

Pricing

ZenML publishes its open-source and Pro Starter prices and quotes Enterprise, as of November 2026.

PlanPriceWhat stands out
Open SourceFreeZenML and Kitaru under Apache 2.0, self-hosted, with pipelines, traces, evals, replay, and community support.
Pro Starter$39 / monthHosted platform for one AI agent and one ML project, with AI engineers, schedules, triggers, and alerts, and a 14-day trial.
Pro EnterpriseCustom quoteMore agents and projects, multi-workspace setup, on-prem or air-gapped deployment, SSO, SCIM, audit logs, and an SLA.

The Field at a Glance

How ZenML compares with other MLOps and experiment tracking vendors we reviewed, by Overall Score and relative typical engagement cost.

6 7 8 9 Overall Score $ $$ $$$ $$$$ Relative typical engagement cost Comet Domino Data Lab Weights and Biases ClearML Union.ai Hopsworks lakeFS Valohai DagsHub ZenML 8.0
ZenML Weights and Biases Union.ai Hopsworks ClearML lakeFS Valohai DagsHub Comet Domino Data Lab

ZenML scores 8.0 in MLOps and experiment tracking, under Weights and Biases (8.9) and Union.ai (8.3), and ahead of Hopsworks (7.8), ClearML (7.7), lakeFS (7.5), Valohai (7.1), DagsHub (6.8), Comet (6.7), and Domino Data Lab (6.5). Relative cost lands lower for this subcategory.

Compared with …

How Well ZenML Fits Each MLOps Use Case

We rated ZenML against the jobs ML and AI teams most often give a pipeline and MLOps framework, drawing on its site, documentation, case studies, and pricing page.

Use caseFitNotes
Portable training and batch pipelinesStrongPipelines run on Kubeflow, Airflow, SageMaker, Vertex AI, Kubernetes, or a laptop without code changes.
Agent testing and improvementStrongKitaru and ZenML Pro import traces from Langfuse, LangSmith, and Braintrust and test prompt fixes against past runs.
Artifact and model lineageStrongEvery pipeline step versions its inputs and outputs, so runs can be traced and reproduced.
Experiment comparison dashboardsMixedRuns and metadata are tracked, but many teams still connect MLflow or Weights & Biases for rich charts.
Managed GPU computeWeakZenML orchestrates a team’s own compute and does not sell GPUs.

Review Excerpts

These excerpts draw on ZenML’s site, public customer statements, and public review sites. Paid reviews and pay-for-play sites were excluded.

What people like

“ZenML has helped us to boost the collaboration and best practices exchange, while keeping the infrastructure burden as low as possible for the MLEs.”

Vladislav Tankov, AI Director, JetBrains
How it's used

“After a benchmark on several solutions, we choose ZenML for its stack flexibility and its incremental process. We started from small local pipelines and gradually created more complex production ones. It was very easy to adopt.”

Clément Depraz, Data Scientist, Brevo
What people don't like

“The open-source core is easy to start with, but we had to wire up our own experiment dashboard and spend time on stack configuration before production runs felt routine.”

ML engineer, midsize software company
Representative buyer view

Where customer reviews couldn't be found, our analysts attempted to give a voice to buyer sentiment, drawing on real information from review sites, forums, and product documentation.

Company Facts

Headquarters
Munich, Germany (remote team)
Founded
2021
Funding
$6.4 million seed led by Point Nine and Crane Venture Partners
Pricing model
Free open source; Pro Starter $39 a month; Enterprise is quoted
Website
zenml.io

Key Leadership

Adam Probst

Co-founder and CEO

He co-founded ZenML and leads the company, now ZenML Labs.

Hamza Tahir

Co-founder and CTO

He co-founded ZenML and leads its engineering and open-source work.

Methodology

We scored ZenML on eight criteria, from experiment tracking to contract fairness, and the Overall Score is their simple average on a 0 to 10 scale. The scores draw on reviews from public review sites and forums, with paid reviews and pay-for-play sites excluded, plus a hands-on look at the product, its pricing, and its documentation and training. Scores can change as we add reviews and product checks.

← Back to MLOps & experiment tracking

Sources

Company facts

Review excerpts