Anomaly Detection — catch metric deviations before they become incidents

Anomaly Detection scans your recent LLM events for statistical deviations and explains what it found in plain English. Select a time range, click Detect Anomalies, and the engine groups events by model, runs Z-score analysis against your baseline, and flags anything that falls outside normal range.

​

Running a scan

1. Open AI Features from the left sidebar.
2. Select a time range: Last 1 hour, Last 24 hours, or Last 7 days.
3. Click Detect Anomalies.

The engine groups events by model before running detection so a spike in one model doesn’t mask healthy behavior in others.

​

What gets detected

​

What anomaly cards show

Each detected anomaly produces a card with:

• Type — LATENCY_SPIKE, ERROR_SPIKE, or COST_SPIKE
• Severity — critical, high, medium, or low
• Description — the specific metric values that triggered the detection
• Affected models — which models are showing the deviation
• AI recommendation — a plain-English suggested next step

​

Natural Language, digest and optimizer tabs

The AI Hub also holds the Natural Language tab, the Overview Digest, the Cost Optimizer and the Prompt Optimizer. On the Natural Language tab, an answer carries the same Verified or Unverified badge as ZespanPilot data answers. If your question can’t be answered from your project’s data, the tab says why instead of returning a best guess. When the AI can’t produce the Overview Digest, a simpler summary is shown until the next load. See the Overview page for details.

​

Incident correlation

Anomalies at medium severity and above are automatically correlated with your Incidents feed. high and critical anomalies also fire your configured alert rules. You can also manually open an incident from any anomaly card using the Open incident button.