Pricing

Start free. Pay for what you use.

Usage is metered in credits, billed per organization — no per-seat fees. Every plan includes the full harness; you only choose how many credits you need. Credits map to real model & tool cost — a typical task runs a few hundred credits.

Free
$0
No credit card
2,500 credits / mo
≈ 3–6 tasks / mo
Lite
$29.90/ mo
$23/mo billed annually
30,000 credits / mo
≈ 38–75 tasks / mo
Starter
Popular
$100/ mo
100,000 credits / mo
≈ 125–250 tasks / mo
Scale
$500/ mo
+10% volume bonus
550,000 credits / mo
≈ 690–1,375 tasks / mo
Agency
$1,000/ mo
+20% volume bonus
1,200,000 credits / mo
≈ 1,500–3,000 tasks / mo

Prices in USD, billed per organization via the app. Need more than Agency? Talk to us about custom volume.

How credit pricing works

A credit is a fixed slice of real model + tool cost, so a plan's credits map directly to usage — you pay for the work a task does, not per seat.

~$0.001
per credit
effective rate on Starter — lower on Scale & Agency
400–800
credits / task
a typical multi-step run
$0
to get going
2,500 free credits / mo · no credit card

Nothing is charged silently: reads are free, credits are spent only when a task runs, and you can cap spend with plan limits and one-time top-ups.

Bring your own model

Already paying for a model deployment? Cut your biggest cost line to $0.

Most of a task's credit spend is LLM tokens. Point every agent at your own OpenAI-compatible endpoint — Azure AI, vLLM, OpenRouter, or self-hosted — and those tokens cost $0 in platform credits: you pay your provider directly. Tool, browser and sandbox calls still meter as usual, so you keep the same harness, approval gate and audit log at a fraction of the run-rate. Available on any paid plan (Lite and up) — a one-time top-up doesn't unlock it.

Every plan includes

Plans differ only by monthly credits — never by features.

  • Orchestrator + 6 specialist agents
  • Per-agent model choice (one AI Gateway)
  • Bring your own model — zero-rated LLM credits (any paid plan)
  • Approval gate on every write — in-app or email
  • 24/7 cron schedules, per timezone
  • Per-account memory + reusable Skills
  • Multi-account / MCC with dry-run preview
  • Google Ads, GA4, BigQuery, Search Console
  • Custom MCP servers + Composio tools
  • Full operation / audit log
  • Organizations with admin / editor / member roles

What's a credit?

A credit is a fixed slice of real model + tool cost. Tasks spend credits on model tokens plus any external tools (search, image/video, browser, sandbox). Cheaper models cost far fewer credits — a typical optimization task runs ≈400–800 credits.

The math: 100,000 credits is $100 (Starter) — about $0.001 per credit, with a lower effective rate on Scale & Agency. So a 400–800-credit task works out to roughly $0.40–$0.80. Choosing a lower-cost model for specialists reduces the credits a task uses. Read the credits guide → Compare model costs →

Pricing FAQ

How does pricing work?
Usage is metered in credits, billed per organization. A credit is a fixed slice of real model and tool cost; a typical task runs ≈400–800 credits. Pick a plan for a monthly credit allotment, or top up any time.
How is a credit priced?
A credit is a fixed slice of real model and tool cost — model tokens plus any external tool calls (search, image/video, browser, sandbox), converted to credits. On Starter, 100,000 credits is $100 (≈ $0.001 per credit), with a lower effective rate on Scale and Agency. A typical task runs ≈400–800 credits.
Do I need a credit card to start?
No. Every organization gets 2,500 free credits a month with no card required, so you can run real tasks before you pay.
Do unused credits roll over?
Plan credits reset each billing period and don't roll over. One-time top-up packs stack on your balance and do not reset.
What happens when I run out of credits?
Runs are blocked at zero credits until the next monthly reset, a plan upgrade, or a top-up — nothing is charged silently.
Can I change plans or cancel?
Yes — manage, upgrade, downgrade or cancel any time on the in-app Billing page. Scale and Agency include a volume bonus (more credits per dollar).
Is every feature included on all plans?
Nearly — the full agent harness, approval gates, schedules, memory, MCP and multi-account are on every plan including Free; plans differ only by monthly credits (and the Scale/Agency volume bonus). The one paid-only add-on is Bring Your Own Model, which needs any paid subscription.
Can I avoid paying for LLM tokens in credits?
Yes — Bring Your Own Model (BYOM), available on any paid plan (Lite and up), points every agent at your own OpenAI-compatible endpoint (Azure AI, vLLM, OpenRouter, self-hosted). LLM tokens on it are zero-rated — you pay your provider directly; tool, browser and sandbox calls still meter in credits. Free-tier orgs need to subscribe to unlock it (a one-time credit top-up doesn't).
Try it on your own account

2,500 free credits a month, no credit card. Reads are free; every write waits for you.

Pricing — Agent Planners