Coming from Frankfurter? Here is the mapping.

The path structure differs, parameters are renamed, and some response fields are new. This is a re-point with small code edits, not a zero-change swap. Once you have the mapping, an afternoon is enough to update your integration.

What changes and what you gain

Frankfurter is open-source and wraps European Central Bank reference data. It works well for weekday reference rates when you do not need a key. exchangerate.dev covers the same ECB series but adds live weekend rates (ECB does not publish on weekends), freshness metadata on every response, and a keyed tier for production quotas.

The core difference is that you move from a public no-auth API to a keyed one. You get a source field that distinguishes live spot from ECB or FRED daily fixes, a market_session field that says whether the FX day is open or in a weekend session, and is_forward_filled on historical calls that tells you when a prior value was carried because no fix was published.

Endpoint mapping

The table below covers the three Frankfurter endpoints most developers use. The operation is the same; the path, parameter names, and response shape change.

Before and after code

Here is a typical Frankfurter call converted to exchangerate.dev. The structure is familiar; the differences are the base URL, the path layout, and the Authorization header.

import requests

# Frankfurter — no auth, ECB reference rates only
resp = requests.get(
    "https://api.frankfurter.app/latest",
    params={"from": "USD", "to": "EUR,GBP"},
).json()

print(resp["rates"]["EUR"])   # 0.xxx

import requests

API = "https://api.exchangerate.dev/v1"
KEY = "exr_live_..."   # free key at https://exchangerate.dev/signup

# Base moves to the path; symbols stays as a query param
resp = requests.get(
    f"{API}/latest/USD",
    headers={"Authorization": f"Bearer {KEY}"},
    params={"symbols": "EUR,GBP"},
    timeout=10,
).json()

print(resp["rates"]["EUR"])        # 0.87531
print(resp["source"])              # ecb_daily
print(resp["market_session"])      # open

The key differences: from becomes the last path segment, to becomes symbols, and Authorization: Bearer exr_live_... is added. You also receive source and market_session in the response without any extra work.

Migrating time-series calls

Frankfurter uses a double-dot range in the path: /{start}..{end}?from=USD. exchangerate.dev uses /v1/range with explicit start_date and end_date query parameters. The double-dot syntax is not supported.

# Before — Frankfurter double-dot range
resp = requests.get(
    "https://api.frankfurter.app/2026-06-10..2026-06-13",
    params={"from": "USD", "to": "EUR,GBP"},
).json()

# After: exchangerate.dev named params
resp = requests.get(
    f"{API}/range",
    headers={"Authorization": f"Bearer {KEY}"},
    params={
        "base":       "USD",
        "symbols":    "EUR,GBP",
        "start_date": "2026-06-10",
        "end_date":   "2026-06-13",
    },
).json()

New freshness fields

Every response from exchangerate.dev includes source and market_session. These do not exist in Frankfurter, so they are additive; you do not need to remove anything, but they are worth reading.

On a weekend date request, Frankfurter returns the Friday close. exchangerate.dev does the same for ecb_daily but sets is_forward_filled: true to say so explicitly. For live rates, market_session: weekend means the rate is being updated even though the ECB is not publishing.