Zapier unifies task metering across AI, code, MCP and SDK with variable rates
Zapier converged its meter: AI steps, code and the SDK now follow the same task-based model and share one task pool with Zaps and MCP, with task rates varying by AI model tier, code runtime and connector type.
One task = one external-app action; AI steps, code and SDK metered separately; Agents Enterprise marked Coming Soon.
AI steps, code, MCP and SDK all draw from a single shared task pool at variable per-action rates (rates vary by AI model tier, code runtime, connector type); Agents Enterprise now live as a contact-sales tier. Headline plan prices unchanged.
Zapier rolled out a unified task-based pricing model across its platform. A new banner on zapier.com/pricing announces that “AI steps, code, and SDK now all follow the same task-based pricing model,” and the pricing page now states that Zap workflows, AI steps, code, MCP and the SDK all draw from the same task allocation “with no separate task budgets by product.”
Crucially, task consumption is now explicitly variable: “Every step in a Zap and every external connector call uses tasks — though rates may vary. AI model tier, code runtime, and connector type each affect how many tasks an action uses.” This is a meaningful evolution from the prior “one task = one external-app action” model toward a complexity-weighted meter, where heavier AI models or longer code runs consume more tasks per action.
Headline plan prices are unchanged: Free ($0/mo, 100 tasks), Professional (from $19.99/mo billed annually; $29.99 monthly), Team (from $69/mo billed annually, 25 users; $103.50 monthly), and Enterprise (contact for pricing). The Agents Enterprise tier, previously labelled “Coming Soon,” is now a live contact-sales plan offering a custom number of activities, org-wide agent sharing, enterprise-grade audit logs and restricted-apps support.
Zapier moved AI steps, Code by Zapier, Zapier MCP and the Zapier SDK onto the same task-based meter — 'AI steps, code, and SDK now all follow the same task-based pricing model.' Tasks are now a single shared account pool ('no separate task budgets by product'), and per-action task rates vary by complexity: a task is 'any successful action that runs,' but AI steps charge by selected model tier, Code by Zapier charges more for extended runtime, each Zapier MCP tool call counts as 2 tasks, and each Lead Router lead counts as 5 tasks. Plan prices and task allowances were unchanged (Free $0/100 tasks, Professional $19.99 annual, Team $69 annual, Enterprise custom); what a 'task' costs changed. Pay-per-task overage now caps at 3× the plan's task limit before runs are held. Agents remain on a separate activity quota.