ZenML
ZenML is a Munich-based open-source MLOps and LLMOps framework company that provides unified orchestration for ML pipelines and AI agents across clouds and environments. It serves data scientists, ML engineers, and enterprise AI teams through an Apache 2.0 core plus a ZenML Pro managed tier.
- Company typePrivate
- Founded2021
- HeadquartersMunich, Germany
- Headcount11–50
- GTM typeB2B
- OfferingSoftware
What ZenML does
ZenML GmbH is a Munich-based open-source MLOps and LLMOps framework company founded in 2021 by Adam Probst and Hamza Tahir. The company builds a unified orchestration layer that enables data scientists, ML engineers, and AI developers to construct reproducible ML pipelines and durable AI-agent workflows that run unchanged across multiple clouds, orchestrators, and environments (Kubernetes, AWS, GCP, Azure, on-prem). Its technology stack centers on a Python SDK with @step/@pipeline decorators, automatic artifact and model versioning, full lineage tracking, smart caching, and a "stack" abstraction layer that decouples pipeline code from underlying infrastructure components such as orchestrators, artifact stores, container registries, and experiment trackers. The framework exposes 60+ native integrations across ML frameworks (PyTorch, TensorFlow, scikit-learn, Hugging Face), orchestrators (Kubeflow, Airflow, SageMaker, Vertex AI, AzureML, Databricks), experiment trackers (MLflow, Weights & Biases, Neptune, Comet), and AI-agent frameworks (LangChain, LangGraph, PydanticAI, OpenAI Agents SDK, Claude Agent SDK, CrewAI, AutoGen).
ZenML operates a hybrid open-source and managed-software business model. The core framework and the Kitaru agent runtime are released under the Apache 2.0 license and distributed self-serve through GitHub, PyPI, and documentation, driving product-led adoption through a 6,200+ star GitHub repository and a 2,900+ member Slack community. Monetization flows through ZenML Pro, a managed control plane offered via SaaS, hybrid, and self-hosted deployment options that adds enterprise-grade capabilities such as SSO, RBAC, team management, resource pools, fair GPU sharing, and a 97% availability SLA. Pricing is quote-based with a "Book a demo" funnel targeting large organizations in insurance (AXA), software (JetBrains), retail (ADEO/Leroy Merlin, IKEA), aerospace (Airbus Defence & Space), automotive (Rivian), telecom (Vodafone), e-commerce (Rohlik), marketing (Brevo), cross-screen media, and academia (Stanford). The company is privately held by ZenML GmbH (Munich HRB 268487), has raised approximately $6.4 million in cumulative funding across a December 2021 seed led by Crane Venture Partners and an October 2023 round led by Point Nine, employs roughly 18 people, and holds SOC2 Type II and ISO 27001 certifications plus CNCF Silver Membership.
ZenML firmographics
Firmographics- Name
- ZenML
- Legal name
- ZenML GmbH
- Website
- https://zenml.io
- Company type
- Private
- Founded year
- 2021
- Operating status
- Operating
- Headcount range
- 11–50 employees
- Short description
- ZenML is a Munich-based open-source MLOps and LLMOps framework company that provides unified orchestration for ML pipelines and AI agents across clouds and environments. It serves data scientists, ML engineers, and enterprise AI teams through an Apache 2.0 core plus a ZenML Pro managed tier.
- Ownership category
- akta.pro rank
ZenML industry classification
Industry- Product category
- MLOps / ML Pipeline Orchestration Software
- NAICS
- Software Publishers (5132), Computer Systems Design and Related Services (54151), Computer Systems Design Services (541512)
- SIC
- Services-Prepackaged Software (7372), Services-Computer Integrated Systems Design (7373)
- akta.pro primary industry
- End-to-End MLOps & ML Platform Suites (HDAAABAA)
- akta.pro secondary industries
- MLOps/LLMOps & Model Lifecycle Management Services (BPAEAHAH), AI/ML Solution Integration & MLOps Enablement (BPAEAAAJ), Kubernetes & Container Platform Management (Private Cloud) (HDABABAC)
Keywords
Where ZenML is headquartered
LocationHeadquarters
- HQ city
- Munich
- HQ country
- Germany
- HQ region
- Europe
Offices2 records
Markets served
ZenML business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations
Revenue model
- Open Source (Apache 2.0): Core ZenML and Kitaru frameworks available under Apache 2.0 license. Free self-hosting with full pipeline orchestration capabilities. Revenue generated through ecosystem adoption leading to Pro conversions.
- ZenML Pro: Managed control plane with Pro-only features including SSO, SLA (97% availability), resource pools, fair GPU sharing, and team management. Upgrade path for scaling teams requiring governance and operational support.
- Enterprise Self-Hosted: Self-hosted deployment option for strict security requirements with complete data sovereignty. Enterprise licensing with support and SLA guarantees.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Freemium | Others | Open Source - Free tier with full framework capabilities |
Go-to-market motion3 records
Distribution channels6 records
Marketing channels11 records
ZenML product offering
Product offeringCore offering
ZenML sells an open-source, Apache 2.0-licensed MLOps and LLMOps framework that lets data scientists and ML engineers build reproducible ML pipelines and deploy AI agents across any cloud, orchestrator, or environment. Its commercial offering, ZenML Pro, adds a managed control plane with SSO, RBAC, resource pools, fair GPU sharing, and SLA-backed availability for enterprise teams, complemented by Kitaru, a framework-agnostic runtime providing durable execution for AI agents with checkpoint, replay, and resume capabilities.
Product overview
ZenML offers a unified ML/AI orchestration platform consisting of three core products: ZenML (open-source ML pipeline orchestration), Kitaru (durable AI agent runtime), and ZenML Pro (managed control plane). ZenML enables users to build ML pipelines that run across any cloud, orchestrator, and environment without rewriting code. Kitaru extends the platform with durable execution capabilities for AI agents, including checkpoint, replay, and resume functionality. ZenML Pro provides a managed version plus enterprise features including SSO, RBAC, and team management. The platform supports 60+ integrations across the AI ecosystem and is designed for teams transitioning from local prototyping to production ML and agent deployments.
Differentiator
Problem solved
Functional benefit
Brands
- Kitaru: The runtime layer underneath the agent stack for durable execution of Python agents — provides checkpoints, replay, resume, and versioned deployments for AI agents. Self-host-first with support for PydanticAI, OpenAI Agents SDK, Claude Agent SDK, LangGraph, and other frameworks.
- ZenML Pro
Products and services
- ZenML Open-source MLOps and LLMOps framework providing unified pipeline orchestration across clouds, orchestrators, and environments, with automatic artifact and model versioning, smart caching, stack abstraction, and 60+ integrations for data scientists and ML engineers building production ML systems.
- Kitaru Framework-agnostic durable runtime layer for AI agents that provides checkpoint, replay, resume, and wait() capabilities, supporting PydanticAI, OpenAI Agents SDK, Claude Agent SDK, LangGraph, and custom Python agent loops for engineers building long-running agent applications.
- ZenML Pro Managed control plane for ZenML and Kitaru workspaces offering SSO, RBAC, resource pools, fair GPU sharing, team management, and a 97% availability SLA, available as SaaS, hybrid, or self-hosted deployment for enterprise teams.
Quantifiable outcome
- 78% faster time-to-market for ML/AI deployments
- +5 more outcomes
Companies that use ZenML
Customer profileNamed customers11 records
Segments4 records
Ideal customer profiles2 records
ZenML technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration72 records
AI capability8 records
Feature6 records
ZenML partnerships and signals
Strategic signalPartnerships
One partnership is on record.
- Cloud Native Computing Foundation (CNCF)ecosystemSilver member of CNCF as of March 2026. Part of 21 new Silver members welcomed during KubeCon + CloudNativeCon Europe. Aligns ZenML with cloud native ecosystem and Kubernetes-native workflows.
Scale indicators7 records
Recent moves4 records
Expansion highlights5 records
ZenML competitors and assessment
Company assessmentDirect peers
- MLflow: Open-source ML lifecycle management platform originally from Databricks. Both ZenML and MLflow target ML pipeline orchestration, experiment tracking, and model versioning, with ZenML positioned as a more orchestration-focused, framework-agnostic alternative.
- Prefect: Python-based workflow orchestration platform with strong overlap to ZenML's pipeline orchestration capabilities. Both compete for data engineering and ML orchestration workloads, with ZenML differentiating through deeper ML-specific abstractions (stacks, artifacts) and Prefect through general-purpose orchestration.
- Dagster: Data and ML orchestration framework with asset-centric abstractions. Competes with ZenML for ML pipeline workloads, with both offering open-source cores plus managed enterprise tiers and Python-native developer experiences.
- Metaflow: Originally Netflix's ML platform, now an open-source framework for production ML workflows. Direct peer to ZenML in the ML pipeline orchestration space, with comparable Python DSL and cloud-deployment model.
- Flyte: Kubernetes-native workflow orchestration platform originally built at Lyft. Directly comparable to ZenML's Kubeflow orchestrator story, with both targeting containerized, production-grade ML pipelines on Kubernetes.
- Kubeflow: CNCF incubating project for ML on Kubernetes. Direct peer in the Kubernetes-native MLOps space and a target orchestrator for ZenML — the two are complementary rather than strictly competitive, but both chase the same enterprise ML platform budget.
Broad incumbents
- Databricks: Unified data + AI platform including MLflow, Databricks Workflows, and Mosaic AI. Operates broadly in the same ML/AI infrastructure space as ZenML but with a much larger portfolio, deeper enterprise footprint, and bundled data lakehouse economics.
- AWS SageMaker: Amazon's end-to-end ML platform spanning training, deployment, MLOps, and now Bedrock-integrated agent workflows. A target orchestrator for ZenML and a broad incumbent competing for the same ML infrastructure budgets.
- Google Vertex AI: Google Cloud's unified ML platform including pipelines, model registry, and the Agent Engine / Agent Development Kit. Direct ZenML integration partner and broad incumbent in the same enterprise ML/agent orchestration category.
Emerging players
- LangChain: LLM application framework with LangSmith observability and LangGraph for stateful agent workflows. Partial overlap with ZenML/Kitaru in the AI agent and LLM workflow orchestration space, particularly relevant for the AI agent customer segment ZenML targets.
Market position
Strengths5 records
Weaknesses4 records
Competitive moat4 records
Key risks5 records
Key highlights7 records
Customer concentration
ZenML social profiles
Digital presenceZenML compliance and trust
Trust signalCompliance2 records
ZenML financial estimates
Financial estimateRevenue estimate
Valuation estimate
ZenML leadership team
Management profileNumber of profiles
Profiles4 records
ZenML funding detail
Funding detailFunding overview
Funding rounds2 records
Investors3 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
ZenML M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about ZenML
What does ZenML do?
ZenML sells an open-source, Apache 2.0-licensed MLOps and LLMOps framework that lets data scientists and ML engineers build reproducible ML pipelines and deploy AI agents across any cloud, orchestrator, or environment. Its commercial offering, ZenML Pro, adds a managed control plane with SSO, RBAC, resource pools, fair GPU sharing, and SLA-backed availability for enterprise teams, complemented by Kitaru, a framework-agnostic runtime providing durable execution for AI agents with checkpoint, replay, and resume capabilities.
Is ZenML a public or private company?
ZenML is a private company. It is classified as venture growth investor backed and is currently operating.
When was ZenML founded?
ZenML was founded in 2021. It employs 11 to 50 people.
Where is ZenML based?
ZenML is headquartered in Munich, Germany, in the Europe region.
How does ZenML make money?
Three revenue lines are on record. Open Source (Apache 2.0) is the primary driver. The others are zenML Pro and enterprise Self-Hosted.
Who are ZenML's main competitors?
Direct peers on record are MLflow, Prefect, Dagster, Metaflow, Flyte and Kubeflow. Broad incumbents are Databricks, AWS SageMaker and Google Vertex AI. LangChain is listed as an emerging player.
Does ZenML have an API?
Yes. ZenML provides a Python SDK that enables developers to build and orchestrate ML pipelines programmatically. The API allows programmatic access to pipeline execution, artifact management, model versioning, and stack configuration. Documentation is available at docs.zenml.io. Kitaru also offers an MCP server for querying and managing executions, deployments, artifacts, stacks, and secret creation through Model Context Protocol tools. Developer documentation is at docs.zenml.io.
What industry is ZenML in?
ZenML's product category is MLOps / ML Pipeline Orchestration Software. Its primary akta.pro industry code is HDAAABAA, End-to-End MLOps & ML Platform Suites, with a secondary code of BPAEAHAH, MLOps/LLMOps & Model Lifecycle Management Services. Its NAICS code is 5132 and its SIC code is 7372.