n8n is an open-source workflow automation platform founded in 2019 by Jan Oberhauser in Berlin, Germany. It provides a visual node-based editor for connecting applications, automating workflows, and building AI agent pipelines. The platform is used by over 100,000 teams globally and is positioned as the developer-friendly alternative to Zapier and Make.com — with a self-hosted Community Edition that is permanently free and unlimited, and a cloud-hosted managed option for teams that prefer to avoid infrastructure management.
The key differentiator from competitors is n8n’s execution-based billing model. One execution equals one complete workflow run regardless of how many steps the workflow contains. A 10-step workflow that runs 100 times costs 100 executions. On Zapier, the same workflow costs 10 × 100 = 1,000 tasks, which at scale pushes Zapier into $400+/month tiers. This makes n8n dramatically cheaper for complex multi-step workflows.
Cloud pricing: Starter at €24/month (or ~$20 on annual billing) for 2,500 executions/month and 5 concurrent workflows. Pro at €60/month (~$50 annual) for 10,000 executions and 20 concurrent workflows. Business at €800/month for 40,000–50,000 executions with SSO, Git version control, and advanced team features. Annual billing saves approximately 17% on all tiers. The Business plan is the largest pricing step and is a significant jump from Pro — most teams considering Business should evaluate whether self-hosting covers their needs first. n8n removed all active workflow limits across every plan in April 2026 — you can now have unlimited active workflows at every tier.
Self-hosted Community Edition is free with unlimited executions, unlimited workflows, and all 400+ integrations. Infrastructure cost for self-hosting is $3–7/month on managed hosting platforms (InstaPods, PikaPods) or $4–$5/month on a raw VPS (Hetzner). For developers comfortable with Docker and basic server management, self-hosting costs 90% less than n8n Cloud for equivalent capability. A free 14-day trial is available for cloud plans (no credit card required for Starter and Pro).
n8n’s AI agent workflows run on the same execution model with no separate AI pricing tier — LLM API calls (OpenAI, Anthropic) are billed separately by the respective API provider. The 2025–2026 update to unlimited active workflows significantly improved the value proposition of cloud plans.