ML Predictions
ML-powered prediction endpoints return statistical estimates trained on SkyLink historical flight data — no local tables required. Both endpoints are available on all RapidAPI plans.
Requirements
x-api-key on every request (direct subscription). Learn more →| Endpoint | Page | When to use |
|---|---|---|
GET /ml/flight-time | Flight Time | Gate-to-gate block time for a city pair, with optional aircraft type and min/max bounds |
GET /carbon/estimate | Carbon Emissions | Per-flight CO₂ estimate using ICAO Doc 9988 methodology |
Choosing an endpoint. Use Flight Time for scheduling, connection buffers, or delay-risk dashboards — returns a point estimate plus min/max bounds for uncertainty bands. Use Carbon Emissions for traveler-facing carbon labels, ESG reporting, or offset flows — supports radiative forcing (RFI) and callsign mode for actual-track distances.
Gate-to-gate block time is elapsed time from pushing back at the departure gate to arriving at the arrival gate, including taxi at both ends.
Error responses
Shared platform errors: Error Handling. Endpoint-specific validation (invalid airport codes, missing route inputs) is documented on Flight Time and Carbon Emissions.
Integration guides
Runnable examples, error handling, and production patterns:
| Guide | Covers |
|---|---|
| Block time | Gate-to-gate estimates and route comparison |
| Carbon labels | CO₂ per passenger, RFI, multi-route ranking |
| Error handling | Validation, missing inputs, disclosure |
| Caching & monitoring | TTLs for predictions and carbon estimates |
Related: Flight Time · Carbon Emissions · Historical ADS-B