Never overspend
on AI.
Set spending limits at every level — organization, team, and individual API key. Rate limits, budget alerts, and automatic top-ups keep you in control without slowing down.
Budget capabilities
Controls at every level.
Per-Key Budget Controls
max spend · daily · monthlySet maximum spend limits on individual API keys. Control daily and monthly budgets independently. When a key hits its limit, requests are blocked instantly — no surprise charges, no overruns.
RPM Rate Limits
requests per minuteCap requests per minute at the key, team, or org level. Prevent runaway loops from burning through credits.
TPM Rate Limits
tokens per minuteControl token throughput to manage costs on high-volume workloads. Set per-key or per-org limits.
Org & Team Budgets
hierarchical enforcementBudget limits cascade from organization to team to individual key. Child limits never exceed parent limits. Complete visibility at every level.
Budget Alerts
Get notified when spending hits configurable thresholds. Never be surprised by a budget breach.
Auto-Topup
Automatic credit replenishment when your balance drops below a threshold. Stay online without manual intervention.
The credit system
Every LLM request uses the reserve-and-settle pattern. Credits are reserved before the call, settled after. No negative balances. Ever.
Reserve
Estimate cost, hold credits
Forward
Send to LLM provider
Settle
Deduct actual cost
Release
Return unused hold
$0.00
Minimum negative
Atomic
All mutations
Ledger
Full audit trail
FAQ
Spend with confidence
Per-key budgets, rate limits, alerts, and auto-topup. Every plan, no paywalls. Full control from day one.