Free Historical Exchange Rate API

Historical exchange rate API free: quick answer

What developers want from a free historical exchange rate API

Query a historical exchange rate by date

The central task is to supply a base currency such as USD, one or more quote currencies such as EUR or CNY, and a past date. The response should identify the observation date because weekends and holidays may use the previous available fixing.

Request a historical exchange rate date range

For charting, reconciliation, and research, users want daily observations between two dates. Check maximum range per request, result ordering, pagination, grouping, and whether missing days are omitted or filled.

Receive a consistent historical exchange rates JSON API response

A dependable response model should represent ISO 4217 currency codes, base and quote currencies, daily reference rates, observation timestamps, time-series dates, quota metadata, and source attribution consistently. Normalize these fields before storing data from more than one provider.

Understand historical exchange rate data before choosing an API

Two APIs can return different values for the same currency pair and date without either response being technically broken. The difference may come from the source, fixing time, market convention, base-currency rule, calculation method, or later revision. Define the rate you need before comparing endpoint features.

Base currency, quote currency, and inversion

A USD/EUR rate commonly means the number of euros for one US dollar, but naming conventions vary. Store the base, quote, value, and provider convention together. If you invert a rate, retain enough decimal precision and record that the value was derived rather than supplied directly.

Reference rates vs market rates

Official daily reference rates are designed for statistical or reporting use. Market feeds may expose bid, ask, midpoint, open, high, low, or close at a specific timestamp. Neither should be presented as the other, and neither automatically includes bank spreads or card fees.

Weekends, holidays, time zones, and revisions

Daily series often omit non-business days or carry the previous observation forward. Providers can also revise past values. Decide whether your system should preserve gaps, use the previous fixing, or reject a missing date, and keep both the requested date and the actual observation date.

Worked non-business-day example: requested date is not observation date

Suppose an application requests USD/EUR for Sunday, 2025-01-05, and the provider returns 0.9700 using the last published fixing from Friday, 2025-01-03. Converting a $10,000 invoice under an approved “previous business-day fixing” policy gives €9,700, but the record must still say requested_date=2025-01-05, observed_date=2025-01-03, and fill_method=previous_fixing. Labeling 0.9700 as a Sunday observation invents a market data point that the source never published.

The same input should produce different storage behavior by use case. An accounting report may carry Friday’s fixing forward when its policy requires a rate for every calendar date. A market-return series should normally keep Saturday and Sunday missing, then join the next genuine observation, because copied values create artificial zero-return days and distort volatility or correlation. Preserve the source publication time and dataset version as well: if Friday’s fixing is later revised to 0.9698, append or version the correction and identify which conversions need recomputation rather than silently changing an audited result.

Currency lifecycle, redenomination, and predecessor codes

Long histories can cross a currency launch, redenomination, peg change, or withdrawal. Confirm whether the API preserves obsolete ISO codes, maps them into a successor currency, or truncates the series. Store the original code and conversion rule; silently rewriting a predecessor currency into its successor creates a false historical comparison.

How to compare free historical currency API options

Do not mix reference rates, executable FX prices, and converters

A central-bank daily reference rate is suitable for many reporting tasks but is not an executable bid or ask. A currency converter may calculate an amount without exposing a historical time series. Keep these intents separate when evaluating APIs.

A practical free-tier checklist

• Confirm historical endpoints are included in the free tier, not only latest rates.
• Record request quotas, rate limits, attribution, caching, and commercial-use terms.
• Check the earliest available date, supported base currencies, and missing-day policy.
• Test error responses for invalid dates, unsupported codes, and quota exhaustion.

Choose the source category before the provider

How to get historical exchange rates via API

Provider-neutral REST request for USD/EUR history

GET /historical?base=USD&symbols=EUR&from=2025-01-01&to=2025-01-31
Accept: application/json

{
  "base": "USD",
  "start_date": "2025-01-01",
  "end_date": "2025-01-31",
  "rates": {
    "2025-01-02": { "EUR": 0.00 }
  }
}

The path and field names are illustrative, not a claim about a specific provider. Do not copy the placeholder rate into production. Adapt the request to the selected API documentation and validate the returned value and observation date.

Validate before storing a daily historical exchange rate

1. Validate ISO 4217 currency codes and the requested date order.
2. Confirm whether the rate means quote units per one base unit.
3. Preserve the provider, source date, retrieval time, and raw response.
4. Handle weekends, holidays, nulls, duplicates, retries, and pagination explicitly.

A production-safe Python ingestion pattern

from decimal import Decimal
from datetime import date
import requests

params = {
    "base": "USD",
    "symbols": "EUR",
    "from": "2025-01-01",
    "to": "2025-01-31",
}
response = requests.get(API_URL, params=params, timeout=20)
response.raise_for_status()
payload = response.json()

for observed_on, values in payload["rates"].items():
    rate = Decimal(str(values["EUR"]))
    assert rate > 0
    save_rate(
        base="USD",
        quote="EUR",
        observed_on=date.fromisoformat(observed_on),
        value=rate,
        provider=PROVIDER_NAME,
        retrieved_at_utc=RETRIEVED_AT,
    )

The field names remain illustrative. Use decimal arithmetic for stored financial values, set explicit timeouts, reject malformed payloads, and make writes idempotent so retries do not create duplicate observations.

Pagination, caching, and retry behavior

• Split long ranges into documented windows and checkpoint the last completed date.
• Use exponential backoff with jitter for retryable responses; do not retry invalid parameters.
• Cache immutable historical observations only after checking whether the provider revises history.
• Store the response version or checksum so changed historical values can be detected.

Historical FX data quality and production checks

Validate completeness and continuity

Build an expected business-day calendar for each series, then classify every gap as an expected holiday, an unavailable fixing, a provider failure, or a processing error. Do not silently turn missing observations into zero.

Detect outliers without inventing replacement rates

Flag impossible values, non-positive rates, abrupt changes, duplicate dates, and inconsistent reciprocal pairs. An alert should trigger review or a second-source comparison; it should not automatically manufacture a replacement value.

Preserve provenance for audit and reconciliation

Monitor the free tier as an external dependency

Track success rate, latency, quota headers, schema changes, stale observations, and license updates. Keep a documented fallback or a recovery export for critical workflows because free access, quotas, and endpoints can change.

How QVeris helps with historical forex rates API discovery

QVeris helps developers and agents discover and inspect callable capabilities. Start with the Alpha Vantage FX Daily tool details, inspect its inputs and outputs, then follow the QVeris REST API reference to connect it. This reduces discovery and schema-inspection work; it does not change a provider’s data, limits, license, or service level.

Free historical exchange rate API FAQ

Authority references