n8n

What is n8n?

Source-available workflow automation platform that can be self-hosted for free. Combines a visual node editor with the ability to drop into JavaScript or Python, and has become the default choice for building AI agent workflows with full data control.

Main features

• Self-hostable (Docker, Kubernetes, npm) with no execution limits on the community edition
• visual node editor plus custom code nodes
• native LangChain integration for AI agent workflows
• vector store nodes (Pinecone, Qdrant, Supabase, PGVector)
• AI Agent node with tool calling
• 1,000+ integrations plus universal HTTP node
• sub-workflows
• queue mode for scale
• MCP client and server nodes.

Practical use cases

• Building RAG pipelines and AI agents with full control over data
• automating internal processes on-premises for compliance
• orchestrating LLM calls across providers with fallback logic
• replacing an expensive Zapier bill with self-hosted infrastructure.

Who is it for?

• Developers, technical operations teams, AI engineers, and organisations with data-residency requirements.

Pricing, free access & limits

Free self-hosted (fair-code licence); paid cloud tiers by execution volume; enterprise self-hosted licence

Plans described in the source

Community Edition free and self-hosted with unlimited workflows and executions (Sustainable Use Licence - free for internal business use, restrictions on reselling as a service). n8n Cloud Starter from roughly EUR 20-24/mo (2,500 executions, 5 active workflows). Pro from roughly EUR 50-60/mo (10,000 executions, 15 active workflows). Enterprise custom, with SSO, LDAP, log streaming, air-gapped deployment and external secrets. Critically, n8n bills per workflow execution, not per step - a 50-step workflow counts as one execution.

Usage limits to consider

Cloud tiers limit executions per month and the number of concurrently active workflows. Self-hosted has no vendor-imposed limits - only your own infrastructure capacity. The Sustainable Use Licence restricts offering n8n itself as a hosted service to third parties.

Strengths & tradeoffs

These points summarize the supplied research, rather than hands-on test results.

Strengths noted in the source

• Per-execution rather than per-step billing makes it dramatically cheaper than Zapier at any real volume
• self-hosting eliminates both cost and data-residency concerns
• native LangChain and vector-store nodes make it the strongest platform here for building real AI agents
• code nodes remove the no-code ceiling.

Limitations to consider

• Genuinely requires technical skill - self-hosting means you own updates, backups, scaling and security
• the licence is fair-code, not true open source, and restricts some commercial uses
• UI is less polished than Make
• fewer prebuilt connectors than Zapier
• community support rather than SLA unless you buy enterprise.

Platforms, languages & integrations

Platforms

• Self-hosted (Docker, Kubernetes, npm, Railway, Render)
• n8n Cloud
• desktop app

Languages

• Interface in English and several community translations

API access

Yes - full REST API, webhooks, CLI, and a node SDK for building custom integrations

Integrations and exports

• 1,000+ nodes including OpenAI, Anthropic, Google, Hugging Face, Pinecone, Qdrant, Supabase, Postgres, MySQL, Slack, Google Workspace, Microsoft 365, plus a universal HTTP node

Evaluation notes

The per-execution billing model is the headline economic advantage - a workflow with 50 nodes costs the same as one with 2. For any team with a developer and a compliance requirement, self-hosted n8n is usually the correct answer in this category.

Explore similar tools

Other tools in AI Automation Tools. Compare their specific features and limits for your task.

Frequently asked questions

Sources & editorial notes

Based on the directory editor’s supplied research. The source does not provide a verification date for this row. Importing a record does not independently verify every product claim.

• Official product website
• Official pricing