Overview
Track tokens and costs by provider; enforce hard monthly limits and prevent overruns.
Problem
Unbounded token consumption surprises finance and strains budgets.
Solution
Usage API aggregates daily tokens/costs; RateLimiter enforces monthly token quotas from InstalledApps.options.
How it works
Set ai_quota_tokens, monitor the usage dashboard, and alert when approaching limits. Over-quota calls fail cleanly with clear messages.
Who is this for
Finance / RevOps
Admins
Engineering Managers
Expected outcomes
- Predictable AI cost
- Provider mix optimized by cost/performance
Key metrics
Overrun incidents
Baseline
3 count/month
Target
0 count/month
Cost variance vs budget
Baseline
25 %
Target
5 %
Gallery
Downloads & templates
Case studies
Finance gains cost predictability
Quotas & routing cut overruns to zero.
Healthcare (no PHI) Enterprise NA
Security impact
- Aggregated usage counters only · PII: none
Compliance
- SOC2
- CCPA (no sale of data)
Availability & next steps
Free
Pro
Enterprise