✈️ Google Flights Scraper — Prices, Schedules & Carbon Emissions

Search Google Flights for one-way, round-trip and multi-city itineraries — real-time prices, full per-leg detail (airline, airport, plane type, exact times), and per-flight carbon-emission data, which no other Google Flights Actor on Apify currently returns. No API key, no login, no headless browser — reads Google's own flight-search data directly, so it's fast and doesn't break when the search-results page gets a visual redesign.

Keywords: google flights api, flight scraper, flight price tracker, carbon emissions flight data, round trip flight search, multi-city flights, flight price monitoring, travel data api.

🌟 Why this one

• Carbon emissions on every flight — carbon_emission_grams + carbon_typical_on_route_grams per result, so you can flag greener options. Not offered by the current top Google Flights Actors on Apify.
• Plane type per leg (e.g. "Boeing 737", "Airbus A350") — useful for seat-comfort or aircraft-based filtering, also not in competing Actors' output.
• Deeper filters than any other Google Flights Actor here: departure/arrival hour windows, min/max layover minutes, specific connecting airports, max total duration, less-emissions-only — on top of the usual price/stops/airline/cabin filters.
• No browser — talks to the same backend Google Flights' own web app uses, directly. Faster, cheaper to run, and nothing to time out waiting for a page to render.
• Round-trip done honestly: Google's own API only itemizes one direction at a time for round trips — rather than hand you a half-detailed row, this Actor runs outbound and return as two full searches and tags each row leg_direction: outbound / return, so every row has complete per-leg detail.

📥 Input

Quick search:

{ "fromAirport": "DEL", "toAirport": "BOM", "date": "2026-12-01", "tripType": "one-way" }

Round-trip:

{ "fromAirport": "JFK", "toAirport": "LHR", "date": "2026-12-01", "returnDate": "2026-12-08", "tripType": "round-trip", "adults": 2 }

Bulk / multi-city, via flightSearches:

{
  "flightSearches": [
    { "fromAirport": "LAX", "toAirport": "JFK", "date": "2026-12-01", "tripType": "one-way" },
    { "tripType": "multi-city", "legs": [
        { "fromAirport": "CDG", "toAirport": "FCO", "date": "2026-12-01" },
        { "fromAirport": "FCO", "toAirport": "CDG", "date": "2026-12-08" }
    ]}
  ]
}

Full filter list (all optional): maxStops, maxPrice, airlines, excludeBasicEconomy, lessEmissionsOnly, earliestDepartureHour/latestDepartureHour, earliestArrivalHour/latestArrivalHour, maxDurationMinutes, connectingAirports, minLayoverMinutes/maxLayoverMinutes, carryOnBags/checkedBags, hideSeparateAndSelfTransfer, currency, language, seatType, adults/children/infantsInSeat/infantsOnLap, maxResultsPerSearch.

📤 Output (one row per flight option)

{
  "success": true,
  "search_index": 0,
  "from_airport": "DEL",
  "to_airport": "BOM",
  "trip_type": "one-way",
  "leg_direction": "outbound",
  "price": 13740,
  "airlines": ["SpiceJet"],
  "stops": 0,
  "total_duration_minutes": 140,
  "carbon_emission_grams": 99000,
  "carbon_typical_on_route_grams": 100000,
  "legs": [{
    "from_airport_code": "DEL", "from_airport_name": "Indira Gandhi International Airport",
    "to_airport_code": "BOM", "to_airport_name": "Chhatrapati Shivaji Maharaj International Airport Mumbai",
    "departure": "2026-12-01 06:20", "arrival": "2026-12-01 08:40",
    "duration_minutes": 140, "plane_type": "Boeing 737"
  }]
}

A search that fails (bad airport code, no route found, etc.) produces one row with success: false and a real error message — it never silently returns nothing, and one bad search in a batch never stops the rest.

🎯 Use cases

• Price monitoring — schedule runs, track a route's price over time, alert on drops.
• Travel apps / aggregators — real-time pricing feed with full per-leg + carbon data.
• Market & fare research — bulk multi-route, multi-date comparison via flightSearches.
• Sustainability-aware booking tools — the only Actor here exposing real carbon-emission figures per flight.

🔌 Use this API from Claude / ChatGPT (MCP)

This Actor works out of the box with Apify's hosted MCP server — no extra setup on this Actor's side. Connect your AI client (Claude, ChatGPT, Cursor, etc.) to https://mcp.apify.com with your Apify token, then ask it to call this Actor by name:

inexhaustible_glass/google-flights-scraper

The MCP server exposes generic tools (search-actors, fetch-actor-details, call-actor, get-dataset-items) that work with any public Actor, this one included — your AI assistant can search for it, read its input schema, run it, and fetch the results, all through the same MCP connection.

❓ FAQ

Does this need a Google account or API key? No — it reads Google Flights' own public flight-search data directly.

How accurate are the prices? They're the live prices Google Flights itself returns at request time — same source data a user would see in their browser, fetched directly instead of through a rendered page.

What about booking links? Every result includes google_flights_search_url — a one-click deep link back to this exact search (route + date) on Google Flights itself, so you can jump straight to booking. A per-airline resolved checkout link isn't available (Google's booking-redirect tokens aren't exposed by the underlying open-source library), but that's also true of every other Google Flights Actor on Apify — none resolve to an airline's own checkout page.

Why does round-trip return two rows per option instead of one combined itinerary row? Because that's honestly how much detail Google's own backend gives back in one step — see "Why this one" above. Each row is still fully real, fully detailed, just tagged by direction.

Does multi-city always work? It works for most route combinations (verified on LAX↔JFK↔LAX, for example), but a small number of specific route/date combos can fail to parse on Google's side. When that happens you get one honest success: false row with the real error — never fake or guessed data.

⚖️ Notes

Reads Google Flights' own publicly-accessible flight-search data. No login, scraping of personal accounts, or private data involved.