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 ≈1,350–1,900 credits.
Subscribe for a monthly credit allotment, or top up credits by the pack or by the amount — no subscription required.
Prices in USD, billed per organization via the app. Need more than Agency? Talk to us about custom volume.
Two ways to buy credits
- A credit allotment that refreshes every billing period
- The lowest price per credit — Scale and Agency add a volume bonus
- Unlocks Bring Your Own Model and the MCP server's tools mode
- Upgrade, downgrade or cancel any time in Billing
- Buy a pack or any amount from 5,000 to 500,000 credits
- No subscription needed — available on every plan, Free included
- Credits never expire and stack on your balance
- Priced above the plan rate: best for a busy month, not steady use
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.
Nothing is charged silently: credits are spent only when a task or schedule runs, the Billing page itemizes them by model and by tool, and you can cap spend with plan limits and one-time top-ups.
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 by monthly credits and how many tasks run at once; the list below is on every plan, Free included.
- 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
- 27 native platforms — Google Ads, Microsoft Advertising, Meta, TikTok, LinkedIn, X Ads, Snapchat, Reddit, Pinterest, DV360, CM360, SA360, Google Analytics, Search Console, BigQuery, Tag Manager, Merchant Center, YouTube, Ad Manager, AdSense, AdMob, Shopify, SHOPLINE, Stripe, HubSpot, GitHub, Klaviyo
- 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 ≈1,350–1,900 credits.
The math: 100,000 credits is $100 (Starter) — about $0.001 per credit, with a lower effective rate on Scale & Agency. So a 1,350–1,900-credit task works out to about $1.35–$1.90 on Starter ($1.13–$1.58 at Agency's rate, $1.62–$2.85 on top-up credits). Most tasks use less; research-heavy and video tasks use more — see credits by task type. The model matters most: the same task on GPT-6 Luna costs a small fraction of what it costs on a frontier model — see credits by model. Read the credits guide → Compare model costs →
Credit top-ups
One-time, no subscription needed. Credits never expire and stack on your balance. Buy them on the in-app Billing page.
Or choose any amount from 5,000 to 500,000 credits — larger amounts get the lower rate.
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 ≈1,350–1,900 credits. Subscribe to a plan for a monthly credit allotment that refreshes every billing period, or buy one-time credit top-ups whenever you need more — with or without a subscription.
- Can I buy credits without a subscription?
- Yes. One-time top-ups are sold by the pack (10,000 credits for $15, 25,000 credits for $35, 50,000 credits for $65, 100,000 credits for $120) or any amount from 5,000 to 500,000 credits, from the in-app Billing page, on any plan including Free. Top-up credits never expire and stack on your balance.
- Top-up or subscription — which is cheaper?
- A subscription. Plan credits cost $0.00083–$0.001 each and refresh every month; top-up credits cost $0.0012–$0.0015 each (larger top-ups get the lower rate) but never expire. Subscribe for steady usage and top up for a busy month.
- 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 ≈1,350–1,900 credits, so it costs about $1.35–$1.90 on Starter, $1.13–$1.58 at Agency's rate and $1.62–$2.85 bought as top-ups. Measured costs by task type are on /docs/credits-by-task.
- How many credits does the same task use on different models?
- It depends mostly on the model. The same median task costs a few dozen credits in model tokens on GPT-6 Luna (one of the cheapest), a few hundred on mid-tier models like Claude Sonnet 5.5 (the default) and thousands on frontier models like Claude Opus — the full table, recomputed daily from each model's real token price, is on /docs/credits-by-model. You choose the model per agent.
- Do I need a credit card to start?
- No. A new Free workspace gets 2,500 credits a month (≈ 1 task), no card required. Free credits go to the first 2 workspaces a person creates in any 30 days, and the free tier shares one monthly capacity — if it runs out, new free-plan runs wait for the 1st (paid and top-up credits are never held).
- Do unused credits roll over?
- Plan credits reset each billing period and don't roll over. One-time top-up credits stack on your balance, are used after the plan's credits, and never expire.
- 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, custom MCP servers, the MCP server's agent mode and multi-account are on every plan including Free; plans differ by monthly credits, how many tasks run at once (2 on Free up to 30 on Agency) and the Scale/Agency volume bonus. Two things need a paid subscription: Bring Your Own Model, and the MCP server's tools mode, where your own agent calls the platform tools directly.
- 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).
- Is there a welcome offer?
- Yes. 20% off your first credit top-up — applied automatically at checkout, any pack or amount, no plan needed.
2,500 free credits a month, no credit card. Reads run at once; by default every write waits for your approval.