Windmill
7.1/10Windmill is an open-source, code-first platform that turns scripts into workflows, AI agent flows and internal apps, mainly for developer and data teams.
Data last updated November 2026 · Our methodology
Overview
Windmill is an open-source platform from Windmill Labs that turns scripts written in TypeScript, Python, Go, Bash, SQL and other languages into scheduled jobs, multi-step flows and internal apps. Its engine is written in Rust and runs on Postgres. AI agent steps can be added to flows. These agents call Windmill scripts or Hub scripts as tools and can connect to HTTP streamable MCP servers. A step can pause for human approval before a tool call. Agents work with OpenAI, Azure OpenAI, Anthropic, Mistral, DeepSeek, Gemini, Groq, OpenRouter, Together AI and custom endpoints. The product suits engineering, data and platform teams that want one place to run automation, with Git sync, a CLI and self-hosting on Docker, Kubernetes or Fargate. Pricing has two main tiers. The free open-source edition gives unlimited executions but caps a deployment at 3 workspaces and 50 users, keeps job runs for 30 days at most, and has no audit logs or SAML. Enterprise starts at about 120 dollars a month and is billed by usage: 20 dollars per builder seat, 10 dollars per non-builder, and 50 dollars per compute unit, where one compute unit is a 2 GB worker. Nonprofits and universities get 40 percent off, and a Pro plan for very small companies exists without a listed price. The main limitation is that Windmill is built for developers. Reviewers say non-coders take a while to get productive, the community is smaller and has fewer prebuilt integrations than n8n, and the AGPL licence needs checking before you embed it in a commercial product.
Our Rating
Who Windmill is best for
Best software for DevelopersGood fit
The product is code-first, with multi-language scripts, a CLI, Git sync and in-browser code editing.
Teams can self-host it, and it has workers, schedules and enterprise SSO and audit logs for internal automation.
Enterprise adds SAML, SCIM, audit logs, autoscaling and 24/7 support, though the vendor is very small.
The open-source edition is free with unlimited executions, and a Pro plan targets seed-stage companies.
Less suited
Reviewers say users without coding skills struggle with the script-based model.
Pros & Cons
Pros
- The open-source edition has unlimited executions and can be self-hosted at no licence cost.
- Scripts in TypeScript, Python, Go, Bash and SQL become flows and internal apps in one tool.
- AI agent steps work with many model providers and MCP servers, and can require human approval before a tool call.
- Usage-based Enterprise pricing is transparent, with listed rates per seat and per compute unit.
- The engine is Rust-based, and reviewers point to strong performance.
Cons
- It is aimed at developers, and a reviewer said a junior developer needed two weeks to become productive.
- It has a smaller community and fewer prebuilt integrations than n8n.
- The AGPL licence needs review before embedding it in commercial products.
- The free edition has no audit logs or SAML and caps job retention at 30 days.
- The vendor is a very small company of around 7 people.
Key Features
Windmill Pricing
Free & Open-source
- Up to 4 groups
- Unlimited executions
- 10 GiB object storage
- SSO for up to 10 users
- Git sync for up to 2 users
- 3 workspaces, up to 50 users
- Community support on Discord
- Job run retention of 30 days or less
- 100 emails per day for email triggers
- Self-host on Docker, Kubernetes or Fargate
Enterprise
- SAML and SCIM
- Commercial licence
- From 120 dollars a month
- Unlimited audit logs and job retention
- Unlimited workspaces, users and groups
- Autoscaling, OpenTelemetry, Private Hub
- Kafka, NATS, SQS, GCP and Azure triggers
- 50 dollars per compute unit (one 2 GB worker)
- 24/7 priority support with a 3-hour response time
- 20 dollars a month per builder, 10 dollars per non-builder
Pro
- Self-hosted only
- For individuals, businesses under 10 employees and 250k dollars revenue, or seed-stage startups
Frequently Asked Questions
Yes. The open-source edition is free with unlimited executions, but it is limited to 3 workspaces and 50 users and keeps job runs for 30 days or less.
Enterprise starts at about 120 dollars a month. Each builder costs 20 dollars a month, each non-builder costs 10 dollars, and each compute unit (a 2 GB worker) costs 50 dollars.
Yes. Both the free and Enterprise editions can be deployed on Docker, Kubernetes or AWS Fargate.
OpenAI, Azure OpenAI, Azure AI Foundry, Anthropic, Mistral, DeepSeek, Google Gemini, Groq, OpenRouter, Together AI and custom endpoints. Agents can also connect to MCP servers.
Yes. Nonprofits and universities get Enterprise at a 40 percent discount.
The free edition has community support on Discord. Enterprise includes 24/7 priority support with a 3-hour response time and a dedicated Slack or Discord channel.
Similar Tools
See all alternatives
Claude Code is Anthropic's agentic coding tool. It reads a codebase, edits files, runs commands and tests, and opens pull requests for developers and engineering teams.
CrewAI is an open-source Python framework and enterprise platform (AMP) for building, deploying, and governing multi-agent AI workflows. It is used by developers and enterprise platform teams.
Onyx, formerly Danswer, is an open-source AI assistant and enterprise search tool that lets staff search and chat across 50+ connected workplace apps with permissions respected.
Ranked near Windmill in AI Software
See Windmill in the listWindmill is #45 in AI Software by our score.
