One flat price. No credits to count.
No token math, no credit ledger, no per-message pricing. Each plan is a flat monthly fee with an included usage allowance and a set of model slots. You see one simple meter — the % of this month's allowance you've used. Start free; upgrade when you keep filling it.
Free
Try the OS
20 lifetime AI actions
- 1 model slot, auto-routed
- Core apps + workspace
- 1 connector
- Community access
Pro
For daily drivers
Base monthly usage allowance
- 2 model slots (1 premium)
- Pick your own models + routing
- Workflow Store + connectors
- Web + Telegram
Pro+
For power users
3× Pro usage
- 3 model slots (2 premium + 1 open)
- Everything in Pro
- More automations
- Longer memory
Max
For builders
6× Pro usage
- 4 model slots (3 premium + 1 open)
- Frontier models always-on
- Studio (media)
- Higher limits
Max+
For teams of one
12× Pro usage
- 5 model slots (3 premium + 2 open)
- Everything in Max
- Early features
- Priority support
Teams
Pooled allowance, shared memory & admin console. $30/seat/mo (min 3).
Enterprise
BYOK, SSO, IP allowlist, data residency & compliance modes. Custom pricing.
How usage works
One meter, not a currency
Your plan shows a single percentage — how much of this month's allowance you've used. Nothing to buy, budget, or convert. It resets at the start of every billing period.
Model slots you control
Each plan includes a number of active model slots. Turn frontier models on for hard work, off for everyday chat — you decide how fast the meter moves.
Hit 100%? Nothing breaks
You keep working on the open models, and can top up or upgrade at any time. No surprise invoice, no overage bill you didn't approve.
Prices shown are launch pricing. Billing turns on with the product — early-access members lock founder rates.