Governance
September 12, 2026
Governing your AI usage means being able to answer three questions at any time: who has access to what, what it costs, and where the limits are. TokenHawk answers all three by construction.
Limits per API key
Each integration — a key — can have its own requests-per-minute budget, its own USD budget, its own time windows. One integration can never starve the others.
Limits per tier
Free, Starter and Pro carry concrete ceilings on throughput, concurrency, optimized tokens and active keys. Business and Enterprise lift those ceilings for usage at scale.
Real-time visibility
The dashboard shows, per key and tenant-wide, actual costs, tokens saved, added latency and error rate — enough to make a call on an integration without guessing.
Budget and alerts
A USD budget can be set on each key, with an alert at 80% and a block at 100% of its consumption — a hard limit set by you, not by TokenHawk.
Governance never relies on trust: every limit is enforced server-side, never left to the caller's discipline.