Gumloop pricing: what you pay, and the open-source route

The three billing shapes, and what drives each

Gumloop bills by credits. Its Pro plan starts at $37 per month with a set of included credits, and model calls, tool calls and paid API pulls all spend them; once they run out, extra usage bills at $0.005 per credit. That means the bill follows the work the agents do rather than the size of the team. Gumloop's own pricing page lists the credit totals.

n8n bills by executions. One execution is a single full run of a workflow, no matter how many steps are inside it, so a long workflow that would be thousands of operations elsewhere is still one execution. The self-hosted Community edition is free, and n8n Cloud starts at 24€ per month for 2,500 executions. n8n's pricing page explains the execution model.

Zapier bills by tasks. Every step and every connector call inside a Zap uses tasks, and the whole account shares one task pool. The Free plan carries 100 tasks a month, and the paid Professional plan starts at $19.99 per month. Zapier's pricing page breaks the tiers down.

Kortix prices by seat plus pooled credits. The Free plan is $0 with 200 credits a month for sandbox compute and one project, and the Team plan is $40 per seat per month with 2,500 credits a month per seat, pooled across the team. Agent compute is billed per second on top, and managed models draw the same pool. Bring your own keys or connect the ChatGPT plan you already pay for and the model spend goes to your provider instead.

What self-hosting does to the cost

Running the open-source system yourself removes the subscription line entirely: self-host is free. What is left is your own infrastructure and, unless you bring your own keys, your model provider's bill. For a team that already runs a server, the recurring cost drops to infrastructure and model calls, with no per-seat fee and no per-task meter. The self-hosted guide walks through the commands, and the alternatives page covers the rest of the field.