Skip to content


Python SDK


You can call the Amadeus APIs using the Python SDK. The Python SDK has been uploaded to the official Python package repository, which makes life easier since you can install the SDK as a regular Python package.


First step is to create the environment. Switch to your project repository and type:

python3 -m venv .venv

A new folder .venv will be created with everything necessary inside. Let's activate the isolated environment with the following command:

source .venv/bin/activate

From now on, all packages installed using pip will be located under .venv/lib and not in your global /usr/lib folder.

Finally, install the Amadeus SDK as follows:

pip install amadeus

You can also add it to your requirements.txt file and install using:

pip install -r requirements.txt

The virtual environment can be disabled by typing:


Your first API call

This tutorial will guide you through the process of creating a simple Python application which calls the Airport and Search API using the Amadeus for Developers Python SDK.


from amadeus import Client, Location, ResponseError

amadeus = Client(

    response = amadeus.reference_data.locations.get(
except ResponseError as error:
  • Once you import the amadeus library, you initialize the client by adding your credentials in the builder method. The library can also be initialized without any parameters when the environment variables AMADEUS_CLIENT_ID and AMADEUS_CLIENT_SECRET are present.
  • The authentication process is handled by the SDK and the access token is renewed every 30 minutes.
  • The SDK uses namespaced methods to create a match between the APIs and the SDK. In this case, the API GET /v1/reference-data/locations?keyword=LON&subType=AIRPORT is implemented as amadeus.reference_data.locations.get(keyword='LON',subType=Location.AIRPORT).

Handling the response

Every API call returns a Response object. If the API call contains a JSON response, it will parse the JSON into the .result attribute. If this data also contains a data key, it will make that available as the .data attribute. The raw body of the response is always available as the .body attribute.

print(response.body) #=> The raw response, as a string
print(response.result) #=> The body parsed as JSON, if the result was parsable
print( #=> The list of locations, extracted from the JSON

Arbitrary API calls

You can call any API not yet supported by the SDK by making arbitrary calls.

For the get endpoints:

amadeus.get('/v2/reference-data/urls/checkin-links', airlineCode='BA')

For the post endpoints:'/v1/shopping/flight-offers/pricing', body)

Video Tutorial

You can also check the video tutorial on how to get started with the Python SDK.

Managing API rate limits

Amadeus Self-Service APIs have rate limits in place to protect against abuse by third parties. You can find Rate limit example in Python using the Amadeus Python SDK here.

Python Async API calls

In a synchronous program, each step is completed before moving on to the next one. However, an asynchronous program may not wait for each step to be completed. Asynchronous functions can pause and allow other functions to run while waiting for a result. This enables concurrent execution and gives the feeling of working on multiple tasks at the same time.

In this guide we are going to show you how to make async API calls in Python to improve the performance of your Python applications.

For all these examples we are going to call the Flight-Checkin Links API.


To follow along with the tutorial you will need the followings:

  • Python version >= 3.8

  • Amadeus for Developers API key and API secret: to get one, create a free developer account and set up your first application in your Workspace.

  • aiohttp: you will use the aiohttp library to make asynchronous API calls. You can install it using the command pip install aiohttp.

  • requests: you will use the requests library for synchronous requests. You can install it using the command pip install requests.

  • amadeus: the Amadeus Pthon SDK. You can install it using the command pip install amadeus.

Async API calls with aiohttp

aiohttp is a Python library for making asynchronous HTTP requests build top of asyncio. The library provides a simple way of making HTTP requests and handling the responses in a non-blocking way.

In the example below you can call the the Amadeus Flight-Checkin link API using the aiohttp library and the code runs in an async way.

import aiohttp
import asyncio
import requests

headers = {"Content-Type": "application/x-www-form-urlencoded"}
data = {"grant_type": "client_credentials",
        "client_id": 'YOUR_AMADEUS_API_KEY',
        "client_secret": 'YOUR_AMADEUS_API_SECRET'}
response =,
access_token = response.json()['access_token']

async def main():
    headers = {'Authorization': 'Bearer' + ' ' + access_token}
    flight_search_endpoint = ''
    parameters = {"airlineCode": 'BA'}

    async with aiohttp.ClientSession() as session:

        for number in range(20):
            async with session.get(flight_search_endpoint,
                            headers=headers) as resp:
                flights = await resp.json()

The above code makes POST request to the Authentication API using the requests library. The returned access token is then used in the headers of following requests to make 20 asyncronous API calls.

Async API calls with thread executor

Since we offer the Python SDK we want to show you how you are able to make async API calls using the SDK. The SDK is built using the requests library which only supports synchronous API calls. This means that when you call an API, your application will block and wait for the response. The solution is to use a thread executor to allow run blocking calls in separate threads, as the example below:

import asyncio
import requests
from amadeus import Client

amadeus = Client(

async def main():

        loop = asyncio.get_event_loop()
        futures = [
            for number in range(20)
        for response in await asyncio.gather(*futures):

OpenAPI Generator

In this tutorial, we'll guide you through the process of making your first API calls using the OpenAPI Generator in Python. To begin, you'll need to retrieve the specification files from the GitHub repository. In this example, you will use the Authorization_v1_swagger_specification.yaml and FlightOffersSearch_v2_swagger_specification.yaml files.

Before getting started make sure you check out how to generate client libraries with the OpenAPI Generator.

Call the Authorization endpoint

You will now learn how to call the POST endpoint in order to get the Amadeus access token.

Open your terminal and generate the Python client with the following command:

docker run --rm \
  -v ${PWD}:/local openapitools/openapi-generator-cli generate \
  -i /local/Authorizaton_v1_swagger_specification.yaml  \
  -g python \
  -o /local/auth
In your local directory you will see the folder auth which contains the generated library.

You can install the library using pip:

pip install openapi-client

Then create a file and add the following code to generate an Amadeus access token.

import openapi_client
from openapi_client.apis.tags import o_auth2_access_token_api
from openapi_client.model.amadeus_o_auth2_token import AmadeusOAuth2Token

auth_configuration = openapi_client.Configuration()
with openapi_client.ApiClient(auth_configuration) as api_client:
    api_instance = o_auth2_access_token_api.OAuth2AccessTokenApi(api_client)

    body = dict(
    api_response = api_instance.oauth2_token(


The code uses the library we have generated to get an OAuth2 access token. With the o_auth2_access_token_api.OAuth2AccessTokenApi() we are able to call the oauth2_token() method.

The body of the request is being created by passing the grant_type, client_id and client_secret to the oauth2_token() method. If you want to know more about how to get the access token check the authorization guide.

Call the Flight Offers Search API

Now let's call the Flight Offers Search API. Since thr OpenAPI Generator works with OAS3 you will have to convert the flight search specification to version 3 using the swagger editor ({:target="_blank"}. To do the convertion, navigate to the top menu and select Edit then Convert to OAS 3.

The process is the same as above. You need to generate the library:

  docker run --rm \
  -v ${PWD}:/local openapitools/openapi-generator-cli generate \
  -i /local/FlightOffersSearch_v2_swagger_specification.yaml \
  -g python \
  -o /local/flights

and then install it in your environment:

pip install openapi-client

Then create a file and add the following code:

import openapi_client
from openapi_client.apis.tags import shopping_api

flight_configuration = openapi_client.Configuration()
api_client = openapi_client.ApiClient(flight_configuration)
api_client.default_headers['Authorization'] = 'Bearer YOUR_ACCESS_TOKEN'

api_instance = shopping_api.ShoppingApi(api_client)

query_params = {
    'originLocationCode': "MAD",
    'destinationLocationCode': "BCN",
    'departureDate': "2023-05-02",
    'adults': 1,
    'max': 2
    api_response = api_instance.get_flight_offers(
except openapi_client.ApiException as e:
    print("Exception: %s\n" % e)

The above code uses the generated library to to search for flight offers. It creates an instance of the shopping_api.ShoppingApi class and setting the default headers to include the access token.

Then it is calling the get_flight_offers() method to make the API request.

Last update: April 15, 2024