Activepieces
Chat, agents, flows, tables and 760+ apps in one workspace, so your team works with AI in the tools it already uses.
By Activepieces · 4.0/5 Value-Position score (estimate)
Positioning guardrails
Best for
- Ops, support and finance teams drowning in recurring manual work they want agents and flows to absorb
- Enterprises that must keep data, runs and credentials inside their own cloud, on-prem or an air-gapped environment
- Companies that need SSO, SCIM, custom roles, audit logs and bring-your-own model keys before AI touches production
- Product teams that want to embed a white-label flow builder and 760+ integrations inside their own SaaS
- Teams replacing Zapier or Make who want an MIT-licensed core and self-hosting as an escape hatch
Ideal size: 10–500 people · Scale-up or enterprise with a security review and automation owners in each department
Not for
- Buyers wanting a fully managed, zero-configuration SaaS with no cloud, on-prem or air-gapped decision to make
- Very small teams: the Plus plan caps at five users and shared projects start only on Team
- Organisations that need HIPAA or FedRAMP assurance, which no available source page claims
- High-volume AI workloads that want flat pricing rather than metered credits
Value metrics scorecard
Time-to-Value
Same day to first flow
~1 days to first production value
Total Cost of Ownership
$2,400/yr
Starts at $0 · Flat monthly plan fee with included run credits; overage at $0.007 per credit; AI steps metered in credits; annual billing saves two months
Implementation Friction
2/5
Engineering + admin effort required
Value-Position score
out of 5 · model estimate
Full cost breakdown
Mandatory implementation fee
None
Seat tiers
Free: 1 user; Plus: up to 5 users; Team: 25 users included then $25/seat/month; Ultimate: unlimited
Add-on costs
- Extra credits: $0.007 per credit on Plus and Team; Ultimate gets volume rates
- AI steps: 2 credits (fast model), 10 (smart), 20 (frontier), 1 per agentic action; bring your own keys to cut each AI step to 1 credit
- White-label Embed builder: from $36,000 a year
- Enterprise control (SCIM, custom RBAC, audit logs, secret managers): custom pricing
Company & support
Who is behind Activepieces, and how your team gets help once it is live.
Company
- Founded
- Not recorded
- Headquarters
- Not recorded
How you get support
- PhoneNot listed
- EmailEnterprise only
- Live chatNot listed
- Support portal / ticketsNot listed
- Community forumAll plans
- Help centre / docsAll plans
- Dedicated account managerEnterprise only
- In person / on-siteNot listed
- Hours
- Not recorded
- Response time
- Not stated
Pricing lists support by humans as Community on Free and Plus, Email on Team, and Dedicated on Ultimate; AI support is shown on every plan. A self-serve documentation site is available to anyone.
“Not listed” means the vendor’s public pages don’t mention that channel, not that it is unavailable. Ask about it during evaluation.
Market position
Where Activepieces sits against its closest alternatives. Pick any two of cost, speed, friction and buyer score, and up to 9 companies to compare.
Quadrant view
Typical annual cost × Time-to-value
The lines cross at the median of the solutions shown, so about half sit on each side of each line. A dashed ring marks an outlier pinned to the edge; hover for its value.
Companies on the chart 6 / 10
- Activepieces
- Geckoboard
- Airbyte
- Better Stack
- Gumloop
- Checkly
Add or change companies
Up to 10 companies including Activepieces. Listed closest first.
Stack fit signal
Compatibility with standard B2B ecosystems.
Ships an official MCP server. Connects to Claude Code, Claude Desktop, ChatGPT connectors and Cursor out of the box.
AI & MCP readiness
What Activepieces ships in AI, and what it asks of your ecosystem.
AI features shipped
Chat turns plain-language requests into flows and agents; agents run multi-step jobs across 760+ apps and any MCP server, pausing for approvals on money or customer actions. AI triage classifies and summarises tickets, invoice intake reads documents, and the knowledge base is searchable by agents. Model choice and provider keys are governed centrally, with every action logged.
Your data & models
- Trains on your data
- Never trains on your data
- Runs on
- OpenAI, Anthropic, Google, Amazon
- AI pricing
- Billed by usage or credits
In your ecosystem
- AI connection
- Official MCP server
- Model key
- Either
- AI usage audit
- Full audit trail
Compliance attestations
* IAPP AIGP certifies individuals, not products. It means named staff hold the credential — not that the platform does.
Bottom line
Activepieces is an open-source (MIT core) AI automation platform that puts agents, flows, tables, chat and 760+ app connectors in one workspace. Teams describe a job in plain language and get an agent or flow; IT keeps control through projects, SSO/SCIM, roles, audit logs, secret managers and bring-your-own model keys. It runs on the vendor's EU cloud in Frankfurt, on your own Docker or Kubernetes infrastructure, or fully air-gapped, with unlimited flows on every plan and metered credits for runs and AI steps.
Frequently asked questions
How quickly can we get a first automation into production?
The vendor says a first flow can ship the same afternoon. The Free plan needs no card, Chat drafts a flow or agent from a plain description, and 760+ pre-built app connectors mean most of the wiring already exists. Self-hosting is a one-command install on Docker or Kubernetes.
What does Activepieces cost as more teams join?
Free is $0 with 1,000 credits a month; Plus is $20 a month for up to five users and 10,000 credits; Team is $200 a month for 25 users and 50,000 credits; Ultimate is custom. Flows are unlimited on every plan and logic inside a flow is free. You pay for runs, with overage at $0.007 per credit, and AI steps cost 2, 10 or 20 credits depending on model class unless you connect your own provider keys.
Can we run it in our own environment, even air-gapped?
Yes. The same enterprise build runs on the vendor's EU cloud in Frankfurt, on your own Docker or Kubernetes infrastructure, or fully air-gapped with no outbound network and the model running on your own hardware. Automations export as plain JSON through the API, and the MIT-licensed core means you can keep running them elsewhere.
How are credentials and agent actions controlled and audited?
An agent references a connection by name and never sees the secret; credentials can stay in your own vault, including HashiCorp Vault, AWS Secrets Manager, CyberArk Conjur and 1Password. Every action, by a person or a model, is recorded in an audit trail that streams to your SIEM, and steps touching money, customers or production can require human approval.
Do you train models on our data?
No. The vendor's security page states that automations, runs, documents and prompts are never training data, for its models or anyone else's. Model input and output are stored in run history under the retention window you set, and inference can point at any OpenAI-compatible endpoint you operate.