What place formats are supported?

City and town names, optionally narrowed by country (two-letter ISO 3166 code) and state, plus US 5-digit ZIP codes. For ambiguous names like Springfield, always pass country and state.

Can it resolve a full street address?

No — this endpoint resolves place names to coordinates, not house numbers. A street address returns an empty list; geocode the city and use the city-center coordinates instead.

Why is the response a list?

Because place names collide — Paris exists in France and in Texas. The API returns every match; take the first, or filter by the country and state fields.

Can I go the other way — coordinates to a city?

Yes. The Reverse Geocoding endpoint takes lat and lon and returns the nearest city, region, and country.

What coordinate system is used?

WGS 84 decimal degrees — the same system GPS and every mapping library use, so the values drop straight into Leaflet, Google Maps, or a database point column.

How to Convert a City to Latitude and Longitude in Python

Prerequisites

You need three things:

Python 3.8 or newer.
The requests library.
A free API Ninjas API key — issued instantly, no credit card. If you already have an account, it's on your profile page.

shell

pip install requests

Step 1: Geocode your first city

The endpoint is /v1/geocoding. Pass a city (a US 5-digit ZIP code works too), and optionally country and state:

geocode.py

import requests

response = requests.get(
    'https://api.api-ninjas.com/v1/geocoding',
    params={'city': 'San Francisco', 'country': 'US'},
    headers={'X-Api-Key': 'YOUR_API_KEY'},
    timeout=10,
)
response.raise_for_status()
results = response.json()

place = results[0]
print(f"{place['latitude']}, {place['longitude']}")
# 37.7790262, -122.4199061

The request on the wire and the full response:

api.api-ninjas.com

GET/v1/geocoding?city=San Francisco&country=US

200 OKapplication/json

[
  {
    "name": "San Francisco",
    "latitude": 37.7790262,
    "longitude": -122.4199061,
    "country": "US",
    "state": "California"
  }
]

Coordinates are WGS 84 decimal degrees — the format every map library and GPS system expects.

Step 2: Handle ambiguous names

The response is a list because city names repeat across the world. The fix is to be specific — pass country, and state where countries reuse names internally. A small helper that returns the best match (or None) keeps the rest of your code clean:

geocode_client.py

import os

import requests

API_URL = 'https://api.api-ninjas.com/v1/geocoding'

session = requests.Session()
session.headers['X-Api-Key'] = os.environ.get('API_NINJAS_KEY', 'YOUR_API_KEY')


def geocode(city: str, country: str | None = None, state: str | None = None) -> dict | None:
    """Return the best match for a place name, or None if nothing matched."""
    params = {'city': city}
    if country:
        params['country'] = country
    if state:
        params['state'] = state

    response = session.get(API_URL, params=params, timeout=10)
    response.raise_for_status()
    results = response.json()
    return results[0] if results else None


place = geocode('San Francisco', country='US')
print(f"{place['name']}: {place['latitude']:.4f}, {place['longitude']:.4f}")

ambiguous.py

# "Paris" alone matches France AND Texas — disambiguate with country/state
paris_fr = geocode('Paris', country='FR')
paris_tx = geocode('Paris', country='US', state='Texas')

print(f"Paris, France: {paris_fr['latitude']:.3f}, {paris_fr['longitude']:.3f}")
print(f"Paris, Texas:  {paris_tx['latitude']:.3f}, {paris_tx['longitude']:.3f}")

Step 3: Geocode a batch

Store lists, office locations, event venues — batch geocoding is a loop with the shared session:

batch.py

places = [('London', 'GB'), ('Paris', 'FR'), ('Tokyo', 'JP'), ('Sydney', 'AU')]

for city, country in places:
    place = geocode(city, country=country)
    if place:
        print(f"{city:10} ({country})  {place['latitude']:>9.4f}, {place['longitude']:>10.4f}")

Geocode once and store the coordinates — place names don't move, so there's no reason to resolve the same city on every page load.

Step 4: Chain into other APIs

Coordinates are rarely the goal — they're the key that unlocks location APIs. The classic chain is geocode-then-weather, both on the same API key:

city_weather.py

def weather_for_city(city: str, country: str) -> dict:
    """Geocode a city, then fetch current weather for its coordinates."""
    place = geocode(city, country=country)
    if not place:
        raise ValueError(f'Could not geocode {city}, {country}')

    response = session.get(
        'https://api.api-ninjas.com/v1/weather',
        params={'lat': place['latitude'], 'lon': place['longitude']},
        timeout=10,
    )
    response.raise_for_status()
    return response.json()


weather = weather_for_city('Reykjavik', 'IS')
print(f"Reykjavik: {weather['temp']}°C, wind {weather['wind_speed']} m/s")

The same two-step works for the Timezone API and Air Quality API — any endpoint that takes lat/lon.

Step 5: Handle empty results

An unknown or misspelled place is not an HTTP error — the API returns an empty list. The helper from Step 2 turns that into None; always check it before using the result:

missing.py

place = geocode('Atlantis')
if place is None:
    print('No match found — check the spelling or add country/state')

Where to go from here

For the reverse direction — coordinates back to a place — use Reverse Geocoding. And if your input is an IP address rather than a city name, the IP Lookup API jumps straight to coordinates. The complete script is below.