Usage & Metrics

Monitor your LLM usage, costs, and performance across all models and agents.

Dashboard

Access the usage dashboard at console.zihin.ai > Observability > Usage.

Available Metrics

Metric	Description
Total Requests	Number of API calls
Total Tokens	Input + output tokens consumed
Total Cost	Aggregated cost across all providers
Average Latency	Mean response time
Success Rate	Percentage of successful calls
Top Models	Most used models by volume
Cost by Provider	Cost breakdown per provider

Telemetry API

Access metrics programmatically:

GET https://llm.zihin.ai/api/telemetry/metrics?period=7d

{
  "totalRequests": 15420,
  "totalTokens": 2500000,
  "totalCost": 45.30,
  "averageLatency": 650,
  "successRate": 99.2
}