█

Anomaly Detection API

A fast, scalable API designed to detect outliers in time-series data. Optimized for MLOps pipelines and real-time monitoring.

View on RapidAPI Try the Demo ↓

Quick Start (cURL)

Send your time-series array to the inference endpoint to receive anomalous indices along with confidence scores.

curl --request POST \
  --url https://anomaly-detection.p.rapidapi.com/v1/detect \
  --header 'Content-Type: application/json' \
  --header 'X-RapidAPI-Key: YOUR_API_KEY' \
  --data '{
    "data": [10.2, 11.5, 10.8, 10.9, 100.5, 11.1, 10.4],
    "sensitivity": 0.85
}'

Response

{
  "status": "success",
  "anomalies": [
    {
      "index": 4,
      "value": 100.5,
      "confidence": 0.992
    }
  ],
  "processing_time_ms": 12.4
}

Interactive Demo

This interactive Marimo notebook simulates the API's underlying algorithm (Isolation Forest) directly in your browser using WebAssembly. No data is sent to a server. It allows you to visualize how the engine classifies outliers under different noise levels.

Interactive WASM Demo. Adjust sliders to tweak parameters.