Stack
n8n
An automation platform you can self-host for free or run in the cloud, with support for AI-agent workflows and custom code.
What it actually does
n8n builds automations as node graphs on a canvas, with native LLM nodes and code steps for cases that no-code tools cannot express. It bills per workflow execution rather than per step, so a 20-node workflow costs the same as a 2-node one.
The Community Edition is source-available and free to self-host with no execution limits, which is the main draw: technical founders run unlimited automation on a small VPS. The trade is that they own updates, backups, and uptime.
There is no free cloud tier. Cloud plans start around EUR 20/mo (roughly $24) for 2,500 executions, quoted in euros, with monthly billing running about 17% higher than annual.
n8n vs its main rivals
The tools people actually weigh against n8n: Zapier, Make, Pabbly Connect, Activepieces. Same criteria for every column, including where n8n loses.
| n8n |  Zapier |  Make | AActivepieces | |
|---|---|---|---|---|
Pricing & access | ||||
| Free tier | YesSelf-host free | Yes100 tasks/mo | Yes1,000 credits/mo | YesSelf-host free |
| Starting price | $0 self-host / ~$24/mo cloud | $19.99/mo | $9/mo | $0 self-host |
| Free cloud tier | NoNo | YesYes | YesYes | YesYes |
Platform | ||||
| Self-host / open source | YesSource-available | NoNo | NoNo | YesMIT license |
| Managed uptime / no maintenance | NoCloud only | YesYes | YesYes | NoCloud only |
Scaling & limits | ||||
| Billing unit | Per execution | Per task (per step) | Per credit/op | Per task |
| App integrations | ~1,000+ | 7,000+ | ~2,000 | ~400+ |
Workflow | ||||
| AI-agent / LLM nodes | YesNative | YesYes | YesYes | YesYes |
| Ease of setup (no ops) | NoTechnical | YesEasiest | YesModerate | NoTechnical |
Developer & API | ||||
| Custom code steps | YesJS/Python | YesPaid | YesYes | YesYes |
| Webhooks | Free | Paid | Free | Free |
- Self-hosted Community Edition is free with no execution limits and runs on a small VPS
- Billing is per execution, not per step, so a 20-node workflow costs the same as a 2-node one
- Native LLM nodes and code steps support AI-agent workflows
- No free cloud tier; cloud starts around $24/mo (EUR 20) for only 2,500 executions
- Self-hosting means you own updates, backups, and uptime, an hour or two a month
- Smaller integration catalog (~1,000) than Zapier or Make, so some apps need custom HTTP calls
Pay less for it
5 ways foundFree with unlimited executions; only cost is a small VPS (about $5/mo)
Cloud monthly billing runs about 17% higher than annual
For complex multi-step workflows, per-execution billing beats Zapier's per-task and Make's per-operation models
If you already run a server, adding n8n via Docker costs nothing extra
Make — If you want a hosted tool with a real free tier and no server to maintain, Make's $9 plan covers most non-technical automation
StackTracker tracks what you actually pay for n8n and every other tool, flags overpayment, and shows the dollars you would save by switching.
Is it the right tool for you
- You are comfortable with Docker and want unlimited automation for the cost of a small VPS
- You build AI-agent or code-heavy workflows
- Your workflows have many steps, where per-execution billing wins
- You do not want to run and maintain a server → use Make
- You need a very long-tail integration and fast setup → use Zapier
Track what n8n and the rest of your stack cost
StackTracker adds up every subscription, plus your hours, so you see the real number.
Prices and limits last verified 2026-07-20.