N8n is a fair-code licensed workflow automation tool that combines AI features with business process automation. You build visually, go deep with code, and deploy on your infrastructure or n8n’s. Pricing counts monthly workflow executions, not users. Starter is 20€ /mo, Pro is 50€ /mo, and Business is 667€ /mo, each billed annually. Without a license key, a self-hosted install runs as the free Community edition.
Official website · n8n.io
N8n fits a technical team that wants a canvas, JavaScript or Python inside a workflow, and a choice between n8n Cloud and a host it controls. The billable unit is a full run from start to finish. Starter’s public cap is 2.5K runs a month; Pro’s card says 10K; Business’s card says 40K. The Cloud trial needs no card and is not the unlicensed Community edition.
Build visually, go deep with code, connect to anything. Every step of your agents' reasoning, traceable on the canvas. Deploy on your infrastructure or ours. You write JavaScript or Python anywhere in your workflow, and you see inputs and outputs next to each step. You can re-run single steps, not your entire workflow, or replay and mock data. Use pre-built nodes for common apps. Custom API connections for everything else. The pricing table also lists code steps in workflows (JS/Python), webhook triggers, and control of n8n via API. Bash scripts and custom nodes are marked available in self-hosted.
Official documentationConnect any model. Inspect every decision. Keep humans in the loop. The homepage says you can handle multi-agent setups and RAG systems and swap cloud or offline models without rebuilding your workflows. Structured inputs and outputs control data to and from AI steps. You can inspect prompt, response, and what happened next. The pricing table names an AI Agent node, MCP triggers, and human approval for tool calls. Each plan comes with a monthly allowance of Assistant credits: 1,600 on Starter and 4,000 or 9,600 on Pro, depending on your plan size. Unused credits do not roll over. If you bring your own key (BYOK), the page says no credit limits apply.
Official documentationAll plans include unlimited users & workflows and every integration. Pricing based on monthly workflow executions, regardless of complexity. You pay for a run from start to finish, not per step. Starter is 20€ /mo, billed annually, with 2.5K workflow executions with unlimited steps, 1 shared project, and 5 concurrent executions. Pro is 50€ /mo, billed annually, with 10K workflow executions with unlimited steps. The card says up to 50 concurrent executions. Business is 667€ /mo, billed annually, with 40K workflow executions with unlimited steps. It says self-hosted only and also includes “Self-hosted Hosted by n8n?”. It adds SSO, LDAP, and Git version control. Enterprise is contact sales, with 200+ concurrent executions and dedicated support with SLA.
Official documentationYou can self-host n8n on your own infrastructure, on-premises, or in a private cloud, using Docker Compose, one-line setup, or other deployment methods. The short command is curl -fsSL https://get.n8n.io | sh. It requires Docker on Linux or macOS (or WSL on Windows). Without a license key, n8n runs as the free Community edition. Adding a Business or Enterprise license key enables those editions. A standard self-hosted version is available on GitHub. The docs call the product fair-code licensed.
Official documentationFollows the homepage’s IT Ops example: onboard a new employee. Use a pre-built node where the app exists and a custom API call where it does not. Write JavaScript or Python only in the step that needs it, then re-run that step instead of the whole flow. One finished pass is one execution, however many nodes it has. On Starter that draws from 2.5K workflow executions a month.
A suggested agent path uses the homepage rules: connect any model, inspect every decision, and keep humans in the loop. The pricing table lists human approval for tool calls beside the AI Agent node. You can read the prompt, the response, and what happened next. The homepage also says you can swap a cloud model for an offline one without rebuilding the workflow. If you are not bringing your own key, Assistant credits refresh monthly and do not roll over.
For enriching incident tickets puts the instance on infrastructure you control. The host docs allow on-premises or a private cloud. The one-line command needs Docker on Linux or macOS, or WSL on Windows. Without a license key that install is the free Community edition. SSO, LDAP, and Git version control are on the Business card at 667€ /mo, billed annually. Do not assume the unlicensed edition includes them.
Start the n8n Cloud trial with no credit card, or run the Docker one-line setup. Without a license key the self-hosted app is the Community edition.
Add pre-built nodes or a custom API connection. Write JavaScript or Python in a step when the node is not enough.
Inspect inputs and outputs beside the step. Re-run a single step, or mock data, before you turn the workflow on.
A run from start to finish is one execution. Compare that volume with 2.5K, 10K, or 40K on the annual cards before you leave the trial.
Public annual rates. All plans include unlimited users and workflows and every integration. You pay for a full execution, not each step. A separate month-to-month dollar amount was not verified here. Ask the vendor before you treat the yearly price as the only rate. The Cloud trial needs no credit card. The Business trial does.
Run the core product on your own host.
Not verified.
Self-hosted, no license key.
First hosted workflows and agents.
20€.
/mo, billed annually.
Solo builders and small production teams.
50€.
/mo, billed annually.
Companies under 100 employees that need scale.
667€.
/mo, billed annually.
Strict compliance and a custom execution volume.
Not verified.
Contact Sales.
A Cloud trial includes Pro-level features with 1,000 executions, 5 concurrent executions, a 180-second timeout, and 800 Assistant credits plus 100 AI credits. It excludes admin roles, workflow diffs, and the n8n public API. A Business trial is 14 days and requires a card. Dollar figures on review pages were not used. The compare table’s max execution duration is 5 min on Starter and 40 min on Pro and Enterprise, which is separate from the trial’s 180-second timeout.
You are comparing the catalog’s Make product.
Make shares the Automation category and adds an e-commerce tag. Choose it when you want that product rather than n8n’s self-hosted license key, MCP and agent nodes, and a bill for full workflow runs. Make’s own limits are not copied here.
The job is Parabola’s data flows, not this canvas.
Parabola is another Automation entry. Choose it when the work is that product’s data flows rather than n8n’s documented mix of a visual canvas, code steps, AI approvals, and a Community edition you host.
Without a license key, self-hosted n8n runs as the free Community edition, and a copy is on GitHub. The Cloud trial needs no card and is capped at 1,000 executions, 5 concurrent executions, and a 180-second timeout. The Business trial is 14 days and requires a card.
SourceStarter is 20€ /mo, billed annually, for 2.5K workflow executions. Pro is 50€ /mo, billed annually, for 10K. Business is 667€ /mo, billed annually, for 40K. Enterprise says contact sales. Users and workflows are unlimited on all plans. Annual billing saves 17%, and the month-to-month euro figure is not shown.
SourceYou pay for full executions, not for each step: one run from start to finish, regardless of complexity. A price note says a daily schedule is about 30 or 31 executions a month, and a run every 5 minutes is about 8,600–8,900. Saved-execution storage is a different table row.
SourceEach plan comes with a monthly allowance of Assistant credits: 1,600 on Starter and 4,000 or 9,600 on Pro, depending on your plan size. Trials include 800 credits, which expire when the trial ends. Credits refresh every month and do not roll over. Extra credits can be purchased. Bring your own key and the page says no credit limits apply.
SourceThe short command is curl -fsSL https://get.n8n.io | sh, with Docker on Linux or macOS, or WSL on Windows. Docker Compose is for production with a database. Listed hosts include AWS, Azure, Google Cloud, DigitalOcean, Hetzner, Heroku, and OpenShift. Npm is deprecated from n8n 3.0. A Business or Enterprise license key changes the edition.
SourceThe homepage says deploy on your infrastructure or n8n’s, including on-prem and air-gapped, or managed cloud. Starter and Pro say hosted by n8n. It says self-hosted only and also includes “Hosted by n8n?”. Enterprise says hosted by n8n or self-hosted.
SourceOct 6, 2026 · NavHunt editorial
NavHunt did not run n8n, did not execute a workflow, and did not install the Docker command. This profile restates the n8n homepage, the pricing page, the docs homepage, and the host page reviewed on 6 October 2026. Case-study hours and customer quotations on those pages remain vendor statements. This description is based on n8n's official materials reviewed on 2026-10-06, including n8n homepage, n8n plans and pricing, n8n docs, Host n8n.
Start a Cloud trial or the Community edition, then compare execution caps before you pick Starter, Pro, or Business.