US Census Geocoder API

Base URL

https://geocoding.geo.census.gov/geocoder

Authentication

No API key, account or attribution requirement. US federal government work, in the public domain.

Rate limit

No published per-IP limit for single lookups. Batch files are capped at 10,000 addresses per submission.

Pricing

Free, with no restriction on storing or redistributing results.

CORS

Not enabled — call it from your server

Official docs

Read the docs

Every request below was executed against the live API on 2026-08-21, and the response shown is the real body it returned — not an illustration.

curl 'https://geocoding.geo.census.gov/geocoder/locations/onelineaddress?address=1600+Pennsylvania+Ave+NW%2C+Washington%2C+DC&benchmark=2020&format=json'

const res = await fetch("https://geocoding.geo.census.gov/geocoder/locations/onelineaddress?address=1600+Pennsylvania+Ave+NW%2C+Washington%2C+DC&benchmark=2020&format=json");
if (!res.ok) throw new Error(`Request failed: ${res.status}`);
const data = await res.json();
console.log(data);

import requests

res = requests.get("https://geocoding.geo.census.gov/geocoder/locations/onelineaddress?address=1600+Pennsylvania+Ave+NW%2C+Washington%2C+DC&benchmark=2020&format=json", timeout=20)
res.raise_for_status()
print(res.json())

{
  "result": {
    "input": {
      "address": {
        "address": "1600 Pennsylvania Ave NW, Washington, DC"
      },
      "benchmark": {
        "isDefault": false,
        "benchmarkDescription": "Public Address Ranges - Census 2020 Benchmark",
        "id": "2020",
        "benchmarkName": "Public_AR_Census2020"
      }
    },
    "addressMatches": [
      {
        "tigerLine": {
          "side": "L",
          "tigerLineId": "76225813"
        },
        "coordinates": {
          "x": -77.03518753691,
          "y": 38.89869893252
        },
        "addressComponents": {
          "zip": "20500",
          "streetName": "PENNSYLVANIA",
          "preType": "",
          "city": "WASHINGTON",
          "preDirection": "",
          "suffixDirection": "NW",
          "fromAddress": "1600",
          "state": "DC",
          "suffixType": "AVE",
          "toAddress": "1698",
          "suffixQualifier": "",
          "preQualifier": ""
        },
        "matchedAddress": "1600 PENNSYLVANIA AVE NW, WASHINGTON, DC, 20500"
      }
    ]
  }
}

result.inputobject

Echo of what you sent, including the benchmark actually applied — worth logging when results look odd.

result.addressMatchesarray

Matches found. An empty array means no match, and that is the normal failure mode.

matchedAddressstring

The standardised address the geocoder matched, in Census formatting.

coordinates.x / coordinates.yfloat

Longitude is `x` and latitude is `y`. The naming is the reverse of the usual habit, so read it carefully.

addressComponents.fromAddress / toAddressstring

The house-number range of the matched street segment. A wide range means a coarse interpolation.

addressComponents.streetName, suffixType, suffixDirectionstring

Parsed pieces of the street name, normalised and uppercased.

tigerLine.tigerLineId / sidestring

Identifier of the road segment matched and which side of it the address falls on.

(geographies endpoint) geographiesobject

State, county, tract and block codes, when you call `/geographies/` rather than `/locations/`.

• Geocode a large US address file with no per-record cost
• Attach census tract and block codes for demographic analysis
• Assign customers to counties or congressional districts
• Standardise messy user-entered US addresses to a consistent format
• Bulk-geocode a research dataset where storing results is a requirement