The Notetakers

Make vs n8n (2026): integrations, pricing and AI graded

n8n is the pick. It grades 3 of 3 on all four rows here, 12 of 12 in total, and Make grades 11 of 12 because self-hosting is not documented on Make's pages. n8n is free as the self-hosted Community edition, and n8n Cloud starts at 20€ a month on Starter, billed annually, for 2,500 executions. Make shelves first on AI agents, which it offers on every plan, Free included.

What changed: Nov 3, 2026: first published; Make and n8n graded on four rows.

Make vs n8n: n8n is the pick for 2026

n8n is the pick under this page's rule: the higher total grade across the four rows. Both tools reach grade 3 on three rows. The fourth, self-hosting and API, decides it.

How we grade

Each tool gets a depth grade from 0 to 3 on four capability rows. Each row is then shelved in this order:

  1. Depth grade, highest first.
  2. Plan required, lowest first: Free, then the entry paid plan, then higher plans, then Enterprise or sales only.
  3. Price: the lowest list price for that plan, with its billing period.

A tie after all three steps goes alphabetical, which says nothing about which tool is better. When a vendor page is silent, the entry says "not documented", which never counts as "no". When the plan required has no published price, an entry with a published price shelves ahead of it. The page pick is the tool with the higher total grade across the four rows. Every entry carries its own "Last checked" date, and every claim comes from the vendor's own pages.

Grade Triggers and actions Branching and error handling AI steps and agents Self-hosting and API
0 None None None None
1 Single step only Filters only AI only through an HTTP call to a model provider One of: inbound webhooks, public API, self-hosting
2 Multi-step workflows, no instant or webhook triggers Branching, no built-in error routes AI steps inside workflows, no agents Two of the three
3 Multi-step workflows plus instant or webhook triggers over a native app catalog Branching plus built-in error routes AI agents that pick and call the platform's own tools All three

Make vs n8n at a glance

Shelved by depth grade, then plan required, then price. Last checked: Nov 3, 2026.

Row n8n grade n8n plan required and price Make grade Make plan required and price Shelved first
Triggers and actions 3 Starter, 20€/mo billed annually (2,500 executions) 3 Not documented on Make's webhook help page n8n
Branching and error handling 3 Not documented on n8n's splitting and error handling docs 3 Not documented on Make's router and error handler help pages Make (full tie, alphabetical)
AI steps and agents 3 Starter, 20€/mo billed annually ("every integration" on all plans) 3 Free, $0 (AI Agents "Available on all plans") Make
Self-hosting and API 3 Free, self-hosted Community edition 2 A paid plan for most API endpoints; self-hosting not documented n8n
Total 12 of 12 11 of 12 n8n is the pick
Fact n8n Make
App catalog 2,000+ integrations 3,500+ apps, plus the HTTP app for any service with an API
Free option Community edition, free when self-hosted; Cloud trial with 1,000 executions, no credit card Free plan: 1,000 credits/mo, 1 MB of data storage, 1 data store
Billing unit Execution: one run of the entire workflow, any number of steps; manual editor runs do not count Credit; by default 1 operation equals 1 credit
Over the limit Not documented Scenarios stop until credits are added

The billing units do not compare one to one. n8n counts a whole run as one execution, while Make counts every operation.

n8n

Last checked: Nov 3, 2026.

  • Triggers and actions, grade 3 (Starter, 20€/mo billed annually). n8n lists 2,000+ integrations, and every plan includes all of them. The Webhook node receives data from apps and services when an event occurs, with a 16MB payload limit. Each execution runs "with unlimited steps".
  • Branching and error handling, grade 3 (plan not documented). The IF and Switch nodes turn a single-branch workflow into a multi-branch one. An error workflow starts with the Error Trigger and controls how n8n responds when an execution fails. The Stop And Error node forces a run to fail under conditions you choose.
  • AI steps and agents, grade 3 (Starter, 20€/mo billed annually). The AI Agent node takes a chat model and one or more tools, and "the agent decides which tools to call to complete a task." At least one tool is required. Workflows can connect LLM providers such as OpenAI, Anthropic and Google. For a step-by-step build across six builders, see how to build an AI agent.
  • Self-hosting and API, grade 3 (Free, self-hosted Community edition). You can self-host n8n on your own infrastructure, on-premises or in a private cloud. Without a license key it runs as the free Community edition. The public API covers many of the same tasks as the editor, on every self-hosted edition and on Cloud Starter, Pro and Enterprise, but not on the free trial. The Webhook node covers inbound webhooks.
  • Plans. Starter: 20€/mo billed annually, 2,500 executions, 5 concurrent executions, hosted by n8n. Pro: 50€/mo billed annually, 10,000 executions, hosted by n8n. Business: 667€/mo billed annually, 40,000 executions, self-hosted only, with SSO and LDAP, multiple environments and version control using Git. Enterprise: contact sales, hosted by n8n or self-hosted, 200+ concurrent executions. All plans include unlimited users and workflows.

Best for: teams that want to self-host the workflow engine and call its public API, free on the Community edition.

Limitation: the public API is not on the n8n Cloud free trial, and the Business plan is self-hosted only.

Make

Last checked: Nov 3, 2026.

  • Triggers and actions, grade 3 (plan not documented). Instant triggers carry an INSTANT label in an app's module list. Custom webhooks give you a URL that accepts any data. Make lists 3,500+ apps, and its HTTP app connects "any application or service that provides an API."
  • Branching and error handling, grade 3 (plan not documented). A router splits a scenario into several chains of modules, and a fallback route takes data that fits no other route. Five error handlers cover failures: Skip, Retry, Resume, Commit and Rollback. Rollback stops the run and reverts changes. Commit stops it and saves what was processed.
  • AI steps and agents, grade 3 (Free, $0). AI Agents are "Available on all plans." An agent can use three kinds of tools: module tools, scenarios and MCP tools.
  • Self-hosting and API, grade 2 (a paid plan for most API endpoints). Webhooks queue up to 667 items for every 10,000 credits licensed per month, up to 10,000 items. Make processes up to 300 incoming webhook requests per 10 seconds. The REST API is mostly on paid plans. Self-hosting is not documented on the Make pages checked. The Make MCP server lets clients such as Claude and ChatGPT reach your account: scenario run scopes on all plans, management scopes on paid plans. For how agents use MCP and APIs, see MCP and API access.
  • Plans. Free, Core, Pro, Teams and Enterprise. Free gives 1,000 credits a month. Credits expire at the end of the term. When they run out, you can upgrade, buy extra credits or turn on auto-purchasing.

Best for: AI agents on a free plan, and error handling with rollback.

Limitation: self-hosting is not documented, and most API endpoints are on paid plans.

Executions vs credits

n8n bills executions. One production run is one execution, whatever the step count or data size. Manual runs from the editor, while you build a workflow, do not count toward the quota.

Make bills credits. By default one operation is one credit. With your own AI provider connection, you pay Make credits for operations and the provider for tokens. With Make's AI Provider, Make bills credits on tokens and operations. When credits run out, scenarios stop until more are added.

If Zapier is also on your list, see Zapier vs Make.

Best for

  • Best for self-hosting and API: n8n (Community edition, free)
  • Best for the largest app catalog: Make (3,500+ apps)
  • Best for AI agents on a free plan: Make (Free, $0)
  • Best for unlimited steps per execution: n8n (Starter, 20€/mo billed annually, 2,500 executions)
  • Best for error handling with rollback: Make (Rollback and Commit handlers)

FAQ

Is n8n better than Make?

Under this page's grade, n8n totals 12 of 12 and Make 11 of 12. The difference is self-hosting, which n8n documents and Make's pages do not.

Is n8n free?

The self-hosted Community edition is free. n8n Cloud offers a trial with 1,000 executions and no credit card.

Can Make be self-hosted?

Self-hosting is not documented on Make's pages. Its published plans run from Free through Core, Pro and Teams to Enterprise.

Is Make free?

Yes. The Free plan gives 1,000 credits a month, 1 MB of data storage and 1 data store.

How much does n8n cost?

Starter is 20€, Pro 50€ and Business 667€ a month, each billed annually. Enterprise is priced through sales.

Do Make and n8n have AI agents?

Yes. Make AI Agents are available on all plans, and n8n's AI Agent node decides which connected tools to call.

What counts as an execution in n8n?

One run of the entire workflow, however many steps it has. Manual runs from the editor do not count.

How many integrations do Make and n8n have?

Make lists 3,500+ apps and n8n 2,000+ integrations. Make's HTTP app connects any service with an API, and n8n's catalog lists an HTTP Request node.

Sources

On this page
Written by
Stacy Wagner
Published
Reading time
8 min