Apache Airflow vs Windmill (2026): which should you pick?

Pick Apache Airflow if

• Fits if you write Python and already run your own servers
• Works for teams that want workflows defined as code with no licence fee

Look elsewhere if

• Skip it if you want a drag-and-drop builder, as workflows are Python only
• Not for you if you have nobody to install, upgrade and secure it
• Tight if you need paid support, because the project lists community help only
• Watch out: managed hosting means a separate vendor and a separate bill

Pick Windmill if

• Fits if your automations are a pile of cron jobs and stray scripts that need one home
• Works for engineering teams who want every change reviewed as a diff before it goes live

Look elsewhere if

• Skip it if nobody on the team writes code, because most steps are written in a programming language
• Tight if you're on Cloud Free, where each extra second of compute counts as another of your 1,000 executions
• Tight if you self-host for free, as Community stops at 3 workspaces and 50 users and keeps run details for 30 days
• Pricey if you self-host and need audit logs or SAML, as that's Enterprise from $120 a month
• Watch out: there's no connector library, and paid self-hosted instances report seat and memory use to Windmill

Apache Airflow

• You install, upgrade, secure and scale it yourself unless you buy a managed service such as Astronomer's Astro.
• The project's site lists a community, meetups and a Slack for help; it offers no paid support plan.
• Defining workflows requires Python knowledge; there is no visual builder in the core project.

Windmill

• Windmill is code-first: scripts and most flow steps are written in a programming language, so it suits engineers more than non-technical users.
• The free self-hosted edition stops at 3 workspaces, 50 users and 10 SSO users, has no audit logs and keeps run details for up to 30 days.
• Cloud Free covers 1,000 executions a month, and an execution is one job of under a second; each extra second of compute counts as another.
• Enterprise and Pro bill at least 2 compute units, and their instances report seat and memory use to Windmill through license telemetry.

Apache Airflow

Pipelines as Python code

Workflows are written in Python, so loops and date logic can generate tasks dynamically, with Jinja templating built in for parameters.

Web interface

A web application shows the status and logs of running and finished tasks, and lets you schedule and manage workflows.

Scalable architecture

A modular design with a message queue lets Airflow spread work over any number of workers.

Provider integrations

Ready-made operators cover Google Cloud, Amazon Web Services, Microsoft Azure and other third-party services, and you can write your own.

AI provider

The Common AI provider reached 1.0 in October 2026, covering agents, sandboxes, human approval and retries for LLM work in pipelines.

Windmill

Scripts in more than 20 languages

A function in TypeScript, Python, Go, Rust, Bash, SQL, PHP or another supported language becomes an API endpoint and a form. Imports are resolved and versions locked at each deploy.

Workflows as a graph or as code

Flows are drawn in a flow editor or written in Python or TypeScript and called as an API endpoint. Retries, approval steps and parallel branches are built in.

Internal apps

An app is a React or Svelte frontend whose backend is your existing scripts and workflows. It can be shared with teams, published publicly or embedded in another product.

AI agents and AI-assisted building

Agents run on any of 12 model providers or your own OpenAI-compatible endpoint, with scripts and MCP servers as their tools. Scripts and flows can also be generated with AI in the UI.

Forks, review and rollback

Changes are made in a fork of the workspace, reviewed as a diff, then deployed to staging and production. Each deployment keeps its version, so an earlier one can be redeployed.

Run history and observability

Every run keeps its inputs, logs and result, and a failure can alert a team on Slack, Teams or by email. Enterprise adds audit logs, queue metrics and OpenTelemetry tracing.

Apache Airflow

• Open source$0

  Apache-2.0 licensed software you install and run yourself; hosting and operations costs are yours.

Windmill

• Free and Open-source$0

  Self-hosted Community Edition with unlimited executions, capped at 3 workspaces, 50 users, 4 groups and 10 SSO users. Community support on Discord.

• Free (Cloud)$0

  Windmill Cloud with 1,000 executions a month, sign-in with Google, GitHub or GitLab, and community support on Discord. No credit card required.

• Team (Cloud)$10

  per seat / month, billed by usage each month

  Set per workspace. A seat covers one developer or two operators and includes 10,000 executions a month. 10 seats at most, with audit logs kept for 7 days.

• Pro (self-hosted)From $48

  per month: $8 per seat plus compute units

  For individuals, seed-stage startups and companies under 10 employees and $250k in revenue. Capped at 10 compute units and 10 SSO users, with no audit logs.

• Enterprise (self-hosted)From $120

  per month: $20 per builder seat, $50 per compute unit

  Audit logs, SAML and SCIM, Git sync, autoscaling, a commercial license and 24/7 priority support with a 3-hour response time. Non-builders count as half a seat.

• Enterprise (Cloud)From $840

  per month: $40 per developer seat plus compute

  A dedicated Windmill Cloud instance with isolated workers and database, unlimited audit log retention and 24/7 priority support.

Apache Airflow

No public discounts listed.

Windmill

• Annual discount

  16% off Enterprise and Pro with annual billing

  With the Annually option selected, the pricing page applies a 16% annual discount to Enterprise and Pro subscriptions, which comes to 10 months' price for the year.

• Nonprofit

  40% off Enterprise for nonprofits

  Windmill gives non-profits the regular Enterprise plan at a 40% discount: $12 per developer seat a month instead of $20 when self-hosted.

• Education

  40% off Enterprise for universities

  Universities get the same 40% discount on the regular Enterprise plan that Windmill gives non-profits.

• Startups

  Pro plan for seed-stage startups

  The self-hosted Pro plan, at $8 per seat against $20 on Enterprise, is open to individuals, seed-stage startups and businesses with fewer than 10 employees and $250k in revenue.