Home / Run Operations & Workflows / Activepieces
Open-source automation where one run costs one credit however many steps it has — and the free plan covers 1,000 runs a month.
The best meter in this job for step-heavy workflows — one credit per run — with the catch that AI steps break the simplicity unless you bring your own keys.
Visit Activepieces ↗ Compare all run operations & workflows agents
Activepieces prices automation the way complex workflows want to be priced: one credit per run, however many steps the run has, on an open-source platform you can also host yourself.
Your workflows are step-heavy or AI-assisted and per-task billing punishes you for it — 1,000 non-AI runs a month are free, and 10,000 cost $16/month billed annually.
You want a finished agent rather than a builder, or you need SSO and projects without jumping to Team at $166/month billed annually.
Workflows are built only from verified capabilities in our listing research — no vendor copy, nothing invented.
Starts A new lead arrives from a form or ad platform
You get One run, one credit — however many steps the routing took
Multi-step lead handling stops costing multi-step money
Starts You describe a recurring job — triage this inbox, draft these summaries
You get Finished work with a run log to review
Approval Gate the steps that touch money, customers or production with a human approval step
Agent work runs inside the same flows and meter as your automations
Starts A flow reaches a step that spends money or contacts a customer
You get Sensitive steps never fire unreviewed
Automation speed without giving up the final say on the steps that matter
Starts You ask Claude or Cursor to run a workflow you built
You get Your automations become tools your assistant can call
The builder and the chat assistant stop being separate systems
Describe the job in plain words; the builder gives you full control when the flow gets hard.
Every piece is free on every plan; connect the two or three tools the flow touches.
Put a human approval step in front of anything that spends, sends or deletes.
Non-AI runs cost 1 credit each; AI steps on included models cost 2–20 — connect your own keys on Plus if AI steps dominate.
Time to first useful result: Under an hour for a first flow · no-code 4/5
Billed per credit, where one credit is one flow run however many non-AI steps it has. AI steps on included models cost extra credits by model class (fast 2, smart 10, frontier 20 per step); on Plus and up, bringing your own AI provider keys drops every AI step to 1 credit. Overage on Plus and Team is $0.007 per credit; on Free, runs past the allowance simply wait until credits refresh.
$0 — free forever, no card
Per workflow run · $0/mo for one person
$16/month billed annually ($20 billed monthly) — 10,000 credits/month
Per workflow run · $16/mo for one person
$166/month billed annually ($200 billed monthly) — 50,000 credits/month
Per workflow run · $166/mo for one person
Custom — credits sized with you, yearly pool
Per workflow run · $null/mo for one person
| Plan | Published price | Solo monthly | Billed by | Limits & overage |
|---|---|---|---|---|
| Free | $0 — free forever, no card | $0/mo | Per workflow run | No pay-as-you-go: runs past the allowance wait until credits refresh · The pricing page describes the Free credit reset inconsistently (comparison table: daily; free-plan FAQ: monthly) — see the price-history note |
| PlusSolo pick | $16/month billed annually ($20 billed monthly) — 10,000 credits/month | $16/mo | Per workflow run | No Projects · No SSO · $0.007 per credit past the allowance |
| Team | $166/month billed annually ($200 billed monthly) — 50,000 credits/month | $166/mo | Per workflow run | SCIM, custom roles, Releases and Git Sync are Ultimate only · $0.007 per credit past the allowance |
| Ultimate | Custom — credits sized with you, yearly pool | $null/mo | Per workflow run | Sales-led pricing · Volume rates (not published) |
AI steps on the included models: a run with several frontier-model steps can burn 20 credits per step, so AI-heavy flows outgrow an allowance far faster than the one-credit-per-run headline suggests — connecting your own AI keys on Plus or above caps every AI step at 1 credit and moves the model cost to your provider bill.
Price record · 2026-11-05: First listing check, 2026-11-05, on activepieces.com/pricing — read in a live rendered browser in BOTH toggle states (Monthly default; Annual 'Save 2 months'): Free $0/1,000 credits, Plus $20 monthly or $16 billed annually for 10,000 credits, Team $200/$166 for 50,000, overage $0.007/credit on Plus and Team, Ultimate custom, Embed from $36k/year. Meter read from the page FAQ: one run is one credit however many steps; AI steps on included models cost 2 (fast) / 10 (smart) / 20 (frontier) credits, or 1 credit with your own keys on Plus and up. DISCLOSED CONTRADICTION on the vendor's own page: the comparison table lists the Free credit reset as 'Daily' and the run-out FAQ says runs resume 'the next day', while the free-plan FAQ says 'a fresh 1,000 credits every month… wait until the next month'. The allowance (1,000 credits) is consistent everywhere; the reset period is not, so this listing states the allowance only.
The same fields on every deep listing in this job, so agents compare field by field. “Not verified” means our research has not confirmed it yet — it never means “doesn't have it.”
Activepieces is the open-source entry in this job, and its meter is the story: one flow run costs one credit however many steps the run has, where Zapier-style rivals charge for every step. A solo founder's build-out is the usual connective tissue — leads routed and enriched, orders synced, reports assembled, agents that triage an inbox or draft a summary across your tools — built three ways: describe it to the built-in chat, assemble it in the visual builder with branching, retries and human approval steps, or drop into TypeScript where a flow needs real code. Agents can be briefed in plain words and run jobs end to end, flows are exposed over MCP so Claude or Cursor can run what you built, and every connector (775 open-source pieces, by the vendor's count) is included on every plan — pricing is by runs, never per connector. You can run it on Activepieces' cloud or self-host the MIT-licensed core on Docker or Kubernetes and leave whenever you like. The honest catches are in the meter's fine print: AI steps on the included models cost extra credits by model class (2, 10 or 20 per step) unless you connect your own AI keys on Plus or above, and the pricing page contradicts itself about whether the free plan's credits refresh daily or monthly — the allowance is 1,000 credits either way, but the reset claim cannot be settled from the page.
Solo Fit stays the public score: the qualitative roll-up of these four published inputs under the How We Rank method. Chat drafting and the visual builder keep it near no-code; the TypeScript option and AI-step credit classes are why ease and clarity sit one notch below the simplest rivals.
The difference: Zapier has the widest catalogue and gentlest start, but bills per task — a 4-step run costs 4 tasks there and 1 credit here.
The difference: Make's canvas is more visual and its entry is $9, but it bills per operation, so multi-step runs burn several operations where Activepieces charges one credit.
The difference: n8n is the other self-host option, billed per execution with a steeper learning curve; Activepieces adds a free cloud tier and chat-built flows.
Nothing at 1,000 non-AI workflow runs a month — the Free plan includes 1,000 credits and one credit is one run however many steps it has. At 10,000 runs a month, Plus costs $16/month billed annually ($20 billed monthly), with overage at $0.007 per credit. AI steps on the included models cost extra credits (2–20 per step by model class) unless you connect your own AI keys on Plus or above.
Yes — $0, no card and no time limit, with unlimited flows and up to 5 users, checked on the pricing page on 2026-11-05. One honest caveat: the same page describes the free credits' reset inconsistently — its comparison table says daily while its free-plan FAQ says monthly — so we state the 1,000-credit allowance, which is consistent everywhere, and not the reset period.
A run is one credit however many steps it has, but AI is the exception: a step on a fast included model costs 2 credits, a smart model 10 and a frontier model 20. On Plus and above you can connect your own AI provider keys, which drops every AI step to 1 credit and moves the model cost to your provider bill.
Yes. The core is MIT-licensed open source and deploys with Docker or Kubernetes on your own infrastructure; the paid cloud adds the governance, agents and support layer. The pricing page lists cloud or on-prem hosting on every plan, and a separate Enterprise control line covers fully air-gapped installs at custom pricing.
Activepieces is not in a published stack yet — stacks are edited buying guides, and memberships are added only when the combination has been worked through.
Compare all run operations & workflows agents → · How we rank → · Get a weekly shortlist →