Plan-based model and feature limits
Your subscription plan determines which LLM models the AI Gateway will route to and which security features are switched on. This article explains what's included at each tier and how to unlock more.
How model access works
Every request through the AI Gateway is checked against a tier-based model allowlist before it is routed. If a model isn't included in your plan, the request is rejected with a clear error rather than silently downgraded — so you always know exactly what you're running.
Model availability increases with your plan:
- ▸Community / Starter — Standard general-purpose models, suited to evaluation and lighter workloads.
- ▸Pro — Everything in Starter, plus additional higher-capability models.
- ▸Enterprise — The broadest model access, including top-tier frontier models.
The exact set of models on each tier can change over time; the Compare Plans matrix in your dashboard always reflects what's currently included.
Want a specific model on a lower plan? Use Bring Your Own Key (BYOK) to connect your own provider credentials. Your key, its scope, and a monthly cost limit are stored encrypted, and requests route through your account.
Feature gating by plan
Beyond models, plans also gate platform features. Higher tiers raise usage ceilings and unlock advanced detection and controls:
- ▸Usage limits — API call volume, number of API keys, team members, and data retention window all scale with your plan.
- ▸Threat detection — Core threat detection is available on every plan; more advanced detection such as shadow and zombie API detection and gRPC monitoring unlocks on higher tiers.
- ▸Access & support — SSO / SAML and priority support are Enterprise features.
Check or change your plan
- 1.Open [Billing overview](/dashboard/billing) to see your current plan, renewal date, and live usage meters.
- 2.Go to [Compare Plans](/dashboard/billing/compare) for a side-by-side feature matrix with monthly and annual pricing.
- 3.Select a plan and choose Upgrade to unlock its models and features immediately.
Usage meters on the Billing overview show how close you are to your API call, team member, and retention limits — a good early signal that it's time to upgrade.
Enterprise plans and custom model allowlists are arranged directly with our team. See Contacting support to get started.