Start free. Scale when it matters.
Every plan includes the full gateway: validation, guardrails, metering, and versioned specifications.
For evaluating AppRamp and small integrations.
- 10,000 requests per month
- 2 MCP servers
- 5 widget generations per month
- Full guardrail pipeline
- Usage dashboard
For side projects going live.
- 100,000 requests per month
- 5 MCP servers
- 30 widget generations per month
- Version rollback
- Email support
For products with real AI traffic.
- 500,000 requests per month
- 15 MCP servers
- 150 widget generations per month
- Version rollback
- Email support
For platforms and heavy workloads.
- 3,000,000 requests per month
- Unlimited MCP servers
- 600 widget generations per month
- Data processing agreement
- Priority support
For workloads beyond Business, with limits set per contract.
- Custom request and widget limits
- Unlimited MCP servers
- Data processing agreement
- Priority support
Prices are per workspace and exclude VAT.
Questions, answered
What counts as a request?
One JSON-RPC call to any of your endpoints — tools/list, tools/call, and the rest all count equally. Failed validation still counts; we did the work of protecting your backend.
What counts as a widget generation?
One AI generation or chat refinement in the Widget Builder. Editing the configuration by hand, re-rendering, previewing, and deploying are free and unlimited on every plan.
What happens when I hit my limit?
Requests are declined with a clear error code (-32006) until the calendar month rolls over or your limit is raised. Your backend is never contacted for declined requests.
Can I change plans later?
Yes. Limits are applied per workspace and can be changed at any time without redeploying anything. Need more than Business? Enterprise limits are set per contract — write to [email protected].