Analytics API

All analytics endpoints require authentication. Data is scoped to your team only.

`GET /analytics/overview`

Overall stats across all traces.

Response:

{
  "total_calls": 1240,
  "avg_latency_ms": 342,
  "total_cost_usd": 2.401234,
  "success_rate": 98.4,
  "error_count": 20
}

`GET /analytics/by-model`

Cost and performance breakdown per AI model.

Response:

[
  {
    "model": "llama-3.3-70b-versatile",
    "call_count": 890,
    "avg_latency_ms": 420,
    "total_cost_usd": 1.820000,
    "success_rate": 99.1
  },
  {
    "model": "llama-3.1-8b-instant",
    "call_count": 350,
    "avg_latency_ms": 180,
    "total_cost_usd": 0.580000,
    "success_rate": 96.8
  }
]

`GET /analytics/by-feature`

Cost and usage breakdown per feature.

Response:

[
  {
    "feature_name": "customer_support",
    "call_count": 720,
    "avg_latency_ms": 380,
    "total_cost_usd": 1.440000,
    "success_rate": 98.6
  },
  {
    "feature_name": "doc_summarizer",
    "call_count": 520,
    "avg_latency_ms": 290,
    "total_cost_usd": 0.960000,
    "success_rate": 97.9
  }
]

`GET /analytics/daily`

Last 7 days of daily usage data. Perfect for trend charts.

Response:

[
  {
    "date": "2026-06-19",
    "label": "Jun 19",
    "call_count": 120,
    "total_cost_usd": 0.240000,
    "avg_latency_ms": 310,
    "error_count": 2
  },
  {
    "date": "2026-06-20",
    "label": "Jun 20",
    "call_count": 340,
    "total_cost_usd": 0.680000,
    "avg_latency_ms": 290,
    "error_count": 5
  }
]

Traces