Features

Analytics

Real-time usage, cost, and performance metrics across all models and applications.

Dashboard metrics

Request volume

Total requests per hour, day, week, or month. Group by model or API key.

Token usage

Prompt and completion tokens broken down by model.

Cost breakdown

USD spend by model, key, and date. Useful for cost attribution.

Latency p50/p95/p99

Response time percentiles by model and endpoint.

Error rate

4xx and 5xx error rates by model. Alerts on anomalies.

Routing decisions

Which model auto-routing selected and why.

API access

Query analytics data programmatically via the analytics API (Pro and above):

bash

curl "https://api.auraon.ai/v1/analytics/usage?from=2026-02-01&to=2026-02-28" \
  -H "Authorization: Bearer ar-your-key"

# Response
{
  "total_requests": 284920,
  "total_tokens": 91234000,
  "total_cost_usd": 43.21,
  "by_model": {
    "claude-opus-4": { "requests": 12300, "cost_usd": 18.40 },
    "gpt-4o-mini":   { "requests": 180000, "cost_usd": 5.40 },
    ...
  }
}

Alerts and notifications

Set up the following alerts in the dashboard:

Daily spend exceeds $X
Error rate exceeds 1% over 5 minutes
Account balance below $10
New API key usage from unknown IP