n8n Pricing 2026: Cloud Plans, Executions and Self-Hosting Cost
Short answer: the n8n Community edition is free to self-host, and n8n Cloud starts at $20/mo billed yearly (Starter, 2,500 workflow executions a month). Pro is $50/mo billed yearly for 10,000 executions, and Business, self-hosted only, is $800/mo billed yearly for 40,000. n8n bills per workflow run, not per step. Verified on n8n's pricing page, October 2026.
Most n8n price write-ups get one of two things wrong. They count steps, when n8n only counts runs, or they mix euro and dollar figures without saying so: n8n shows prices in the visitor's currency, and European visitors see euros. Every price below is the dollar price n8n shows US visitors.
n8n plans at a glance
| Plan | Price billed yearly | Executions a month | Concurrent runs | Hosting | Notes |
|---|---|---|---|---|---|
| Community | $0 | No plan quota | Set by your server | Self-hosted | Free license, you run the server |
| Starter | $20/mo | 2,500 | 5 | n8n Cloud | 1 shared project, 1,600 Assistant credits a month |
| Pro | $50/mo | 10,000 (50,000 tier available) | 20 (50 on the 50,000 tier) | n8n Cloud | 3 shared projects, admin roles, workflow history |
| Business | $800/mo | 40,000 | Set by your servers | Self-hosted only | SSO and LDAP, Git version control, environments |
| Enterprise | Contact sales | Custom | 200+ | Cloud or self-hosted | Dedicated support with SLA, log streaming |
Prices read on n8n.io/pricing on October 28, 2026, in US dollars. n8n says yearly billing saves 17% against month-to-month; the monthly prices for US visitors show only after you switch the page toggle, so check them on n8n's pricing page if you want to pay monthly. Every plan includes unlimited users, unlimited workflows and every integration.
Dupple data: Toolradar, our software directory, evaluates 1,033 workflow automation tools as of October 2026 (Toolradar's workflow automation ranking). n8n's per-run billing is the exception in that crowd, and our cost-per-1,000-executions table below shows why it matters once your workflows get long.
How much does n8n cost in 2026?
For a small team on n8n Cloud, $240 to $600 a year. That is our arithmetic on the yearly prices: Starter at $20 a month comes to $240 a year, and Pro at $50 a month comes to $600. Both are paid for the year upfront. Teams that outgrow Pro's 10,000 executions can move to Pro's 50,000-execution tier, while companies that need SSO, Git-based version control and separate environments move to Business, which only runs on your own servers.
Self-hosting the Community edition costs only your server and your time. n8n's hosting docs say that a self-hosted install without a license key runs as the free Community edition. There is no license fee and no plan quota on that edition.
Business is a different price class. At $800 a month billed yearly, it comes to $9,600 a year (our arithmetic) for 40,000 executions. n8n aims it at companies with fewer than 100 employees, and its Start-up Plan gives companies under 20 employees 50% off Business if they qualify.
What counts as an execution in n8n?
One execution is one complete run of a workflow, however many steps it has. n8n's pricing FAQ puts it this way: "An execution is a single run of your entire workflow. It doesn't matter how many steps are in the workflow or how much data it processes." A two-step workflow and a 20-step workflow each cost one execution per run.
That rule is the reason long workflows are cheap on n8n. Tools that bill per step or per task charge again for every action inside the run; n8n does not. The count that matters for your plan is production executions, the ones started by a trigger or webhook. n8n's concurrency guide adds that manual test runs, sub-workflow runs and error workflow runs do not count toward your concurrency limit.
To estimate your own volume, start from n8n's examples: a daily schedule is 30 or 31 executions a month, a workflow every 5 minutes is about 8,600 to 8,900, and a chatbot uses one execution per agent turn, so a busy chat agent hits limits sooner than a nightly report.
Cost per 1,000 executions
The table runs three monthly volumes a small team might reach, with the plan each one needs.
| Monthly executions | n8n Cloud plan | Cloud cost a year | Cost per 1,000 executions | Self-hosted Community |
|---|---|---|---|---|
| 2,500 | Starter, $20/mo yearly | $240 | $8.00 | Server from $12/mo, plus upkeep |
| 10,000 | Pro, $50/mo yearly | $600 | $5.00 | Server from $12/mo, plus upkeep |
| 40,000 | Business (self-hosted), $800/mo yearly, or Pro's 50,000 tier on Cloud | $9,600 on Business | $20.00 on Business | Server from $24/mo, plus upkeep |
Our arithmetic on prices from n8n pricing and DigitalOcean Droplet pricing, October 28, 2026. Cost per 1,000 = monthly price divided by thousands of executions. Server prices are DigitalOcean Basic Droplets: 2 GiB of memory for $12/mo and 4 GiB for $24/mo.
Pro is the cheapest Cloud plan per execution. Starter costs more per run because it is the entry plan, and Business costs the most because you pay for governance features (SSO, environments, Git), not for volume. If you need 40,000 runs and none of those features, compare Pro's 50,000-execution tier on the pricing page before you sign a Business contract.
How much does self-hosting n8n cost?
The software is free; the server and the upkeep are not. n8n's DigitalOcean setup guide says a basic shared CPU plan is enough for most usage levels, which is why the table uses the $12 Droplet as the floor. For context, n8n's Cloud data guide gives Starter instances 640MiB of RAM, and says n8n alone uses about 180MiB.
The bigger cost is time. n8n's deployment guide lists technical expertise as required for installation and configuration, and maintenance as your responsibility: updates, backups, security patches and uptime. If nobody on your team owns that work, a $20 or $50 Cloud plan is cheaper than an engineer's afternoon each month.
One limit people miss: the Community edition runs under n8n's Sustainable Use License, a fair-code license. It allows use for your own internal business purposes or non-commercial use, and it allows distribution only free of charge for non-commercial purposes. If you plan to resell n8n to clients, read the license FAQ before you build.
n8n Starter vs Pro: which one do I need?
Pick Starter if you run under 2,500 executions a month and one person builds the workflows. It includes 5 concurrent runs, 1 shared project, 1,600 Assistant credits a month and Agents, MCP and AI nodes. Its limits are tight: a 5-minute maximum run time, 7 days of execution logs and 2,500 saved executions.
Pick Pro once workflows go into production for a team. Pro raises concurrency to 20 (50 on the 50,000 tier), the maximum run time to 40 minutes and log retention to 30 days, and adds admin roles, workflow history, 3 shared projects and up to 9,600 Assistant credits. Those limits matter more than the execution count: an AI agent workflow that needs six minutes per run is over Starter's limit no matter how few runs you have.
Both plans include AI models without your own API keys, paid from a prepaid Gateway credits balance, so model usage is a separate line on your bill from the plan price.
What happens when I hit my execution limit?
Starter and Pro are capped and have no overage fees. n8n's pricing table says so plainly, which means a busy month does not produce a surprise invoice; to run more, you move to a bigger plan. Concurrency works differently: runs over your concurrent limit wait in a queue and are processed in order once capacity frees up.
Business and Enterprise keep running past quota. n8n's FAQ says workflows continue without interruption, but overage charges may apply if you do not move to the next usage tier. On Business, n8n prints the overage price as $4,800 per extra 300,000 executions, effective since October 26, 2026, which works out to $16 per 1,000 extra runs (our arithmetic). Enterprise overage is set in your order form.
Is n8n free?
The self-hosted Community edition is free; n8n Cloud is not. Cloud has no free plan, only a 14-day free trial with Pro features, 1,000 executions and 5 concurrent runs, and no credit card. If you do not upgrade by the end of the trial, n8n's trial guide says the trial expires and n8n deletes your instance, so export anything you want to keep. The Business plan has its own 14-day trial, which does need a card.
Is n8n cheaper than Zapier?
For workflows with several steps, yes. Zapier bills a task for every action step, while n8n bills one execution per run.
| Scenario | Zapier | n8n |
|---|---|---|
| 5-step workflow, 1,000 runs a month | 5,000 tasks; Professional starts at $19.99/mo billed yearly for its smallest task tier | 1,000 executions; fits Starter at $20/mo billed yearly |
Our arithmetic on Zapier pricing and n8n pricing, October 28, 2026. Zapier's free plan (100 tasks a month) still wins for very small, short automations. Our full Zapier vs n8n comparison runs 2, 5 and 10-step workflows side by side.
Hidden costs to budget for
- Gateway credits for AI models: prepaid, on top of the plan price, if you use n8n's built-in model access instead of your own API keys.
- Assistant credits beyond your allowance: Starter includes 1,600 a month and unused credits do not roll over; extra credits are bought separately.
- Business overage: $4,800 per extra 300,000 executions once you pass your quota.
- Self-hosting time: updates, backups and patches; n8n's guide lists maintenance as your responsibility.
- Yearly commitment: the prices in this guide assume yearly billing; month-to-month costs more.
For the wider field, our best workflow automation tools roundup ranks n8n against its rivals, best Zapier alternatives covers the switch from per-task billing, best AI workflow automation tools covers AI-first builders, and best free AI automation tools lists zero-budget options. Planning to build n8n workflows for clients? Read how to start an AI automation agency alongside the license terms above.
How we checked these prices
Every price was read on n8n.io/pricing on October 28, 2026, in the US dollar prices n8n shows US visitors, along with n8n's docs on hosting, concurrency, Cloud data limits, the free trial and the license. Server prices come from DigitalOcean's Droplet page the same day. Annual totals and cost per 1,000 executions are our arithmetic.
FAQ
How much does n8n cost per month?
n8n Cloud Starter is $20 a month and Pro is $50 a month, both billed yearly, for 2,500 and 10,000 executions. Business, self-hosted only, is $800 a month billed yearly for 40,000 executions. The Community edition is free to self-host.
Is n8n free to use?
Yes, if you self-host the Community edition: there is no license fee, only your server and maintenance time. n8n Cloud has no free plan, only a 14-day trial with 1,000 executions and no credit card required.
Does n8n charge per step?
No. n8n charges per execution, one full run of a workflow, whatever the number of steps or the amount of data. A 20-step workflow that runs 1,000 times uses 1,000 executions.
What happens if I go over my n8n execution limit?
Starter and Pro are capped with no overage fees, so you move to a larger plan to run more. Business and Enterprise keep running past quota and may pay overage; Business overage is $4,800 per extra 300,000 executions.
Is n8n open source?
n8n describes its license as fair-code. The Community edition is free under the Sustainable Use License, which allows internal business and non-commercial use but limits commercial redistribution. Business and Enterprise features need a paid license.
Which n8n plan is best for a small team?
Pro, for most teams running workflows in production: 10,000 executions, 20 concurrent runs, 40-minute runs, admin roles and workflow history for $50 a month billed yearly. Starter fits one builder with light, short workflows.
Cite this: Dupple, "n8n Pricing 2026: Cloud Plans, Executions and Self-Hosting Cost", November 2026.
Related
- Zapier vs n8n: per-task against per-run billing, with a cost table.
- Best workflow automation tools: the full category roundup.
- n8n on Toolradar: listing details and alternatives.
- Best automation tools: automation beyond workflow builders.
- Best AI agent platforms: if agents are the main job.
- Dupple learning hub: more guides on AI tools for work.
n8n is the cheapest way to run long workflows, as long as you budget for the plan limits or the server time. Want weekly picks like this for AI at work? Try Dupple X →
Some offers on this page may be paid placements or contain affiliate links.
