PRICING Deterministic pricing. No surprises. Allowances and limits come from a single authoritative plan catalog — the same source the entitlement service and pricing engine read.
Free Explore the frontier A generous starter tier for individuals exploring AI-assisted development on their own machine.
▸ 200 AI requests / mo ▸ 0.5M input · 0.1M output tokens / mo ▸ Local model execution · Advanced agents · Priority inference queue · Team workspaces & org controls · SSO / SCIM · Usage-based overage billing · Dedicated support MOST DEPLOYED
Pro Full-throttle development $20.00 /mo
or $192.00/yr — billed annually
Higher allowances, frontier models, and priority throughput for professional developers.
▸ 5,000 AI requests / mo ▸ 20.0M input · 6.0M output tokens / mo ▸ Local model execution ▸ Advanced agents ▸ Priority inference queue · Team workspaces & org controls · SSO / SCIM ▸ Usage-based overage billing · Dedicated support MONTHLY YEARLY −20%
Upgrade to Pro
Enterprise Mission control for teams Custom
negotiated · invoiced
Organizations, centralized billing, custom limits, SSO, and negotiated pricing for companies.
▸ Unlimited AI requests ▸ Unlimited tokens ▸ Local model execution ▸ Advanced agents ▸ Priority inference queue ▸ Team workspaces & org controls ▸ SSO / SCIM ▸ Usage-based overage billing ▸ Dedicated support Overage is billed only where the plan enables it, priced by versioned rules resolved server-side at request time — never by client-reported amounts. Taxes calculated at checkout. Cancel anytime; plans remain active through the paid period.