How far back does the news archive go?#

APITube does not publish a single fixed “N years back” cut-off, because archive depth varies by source, language and topic — a large English wire agency reaches back much further than a niche local site added last month. The reliable way to know how deep the archive is for your query is to sort your results oldest-first and read the earliest article’s date. No API parameter reports a headline depth figure, so you measure it directly instead of trusting a number.

There is no separate “archive” endpoint either: historical and fresh articles come from the same /v1/news/everything search. That means the same trick — sort ascending, take the first result — works on any filtered slice you care about.

Is there a fixed cut-off date for the archive?#

No. The News API exposes no field or parameter that returns a global “earliest date” or a total-years number, and the coverage figures on the marketing site are not a per-query guarantee. What actually determines how far back a result set reaches is your filters: the language, source, country or entity you asked for. A query restricted to one small publisher will bottom out at that publisher’s oldest ingested article; a broad query across many sources reaches back to the oldest article among all of them.

Because depth is query-dependent, the honest answer is a method, not a constant: ask the archive for the oldest article that matches your filters and read its timestamp.

How do I find the earliest article for my query?#

Sort by publication date ascending and request a single article. By default results come back newest-first (sort.by defaults to published_at, sort.order to desc), so flip the order to asc and set per_page=1:

curl "https://api.apitube.io/v1/news/everything?sort.by=published_at&sort.order=asc&per_page=1&api_key=YOUR_API_KEY"

The published_at of that one article is the earliest date the archive currently holds for an unfiltered query. To measure the depth of a specific slice, add your filters before running it — for example, the oldest German-language article, or the oldest article from one domain:

curl "https://api.apitube.io/v1/news/everything?language.code=de&sort.by=published_at&sort.order=asc&per_page=1&api_key=YOUR_API_KEY"

Pass your key as the api_key query parameter or the X-API-Key header. This probe costs the same as any small search and answers the depth question precisely for the exact query you plan to run.

How much history exists in a given period?#

To gauge volume rather than the earliest date, use /v1/news/count with a date range. On a live key, passing a real filter (such as published_at.start / published_at.end) returns the true number of matching articles for that window without downloading them:

curl "https://api.apitube.io/v1/news/count?published_at.start=2023-01-01&published_at.end=2023-12-31&api_key=YOUR_API_KEY"

Two caveats worth knowing: a count with no filters at all returns a large rounded showcase figure rather than a literal database size, and a test key always returns a showcase number instead of the real volume. So to profile how much history exists in a period, use a live key and always include the date range. Stepping the window month by month is a quick way to see how article volume grows over time — see how to count news articles for the endpoint’s details.

Can I read the whole archive at once?#

No — a single response is capped at per_page=250 (a larger value returns ER0171), so any deep window is retrieved across multiple pages. Fix a date range, sort ascending, and increment page until the results run out, then move the window and repeat. The querying mechanics — accepted date formats, pagination, and the plan-based page limits — are covered in full in can I access historical news archives? and the parameter itself in how to filter news by date range. The complete date syntax also lives in the official parameters reference.

Common Questions#

• Does APITube have a separate archive endpoint?
• Why isn’t there a single “N years” number?
• Can I reach older articles on the Free plan?

Does APITube have a separate archive endpoint?#

No. Historical articles and fresh articles are served by the same /v1/news/everything endpoint — what makes a request “historical” is the date filter you add (published_at.start / published_at.end), not a different URL. Every other filter (language, source, entity, country) combines with the date range in the same request, so you narrow the archive exactly like you narrow recent news.

Why isn’t there a single “N years” number?#

Because coverage depth is not uniform. Different sources were added at different times, and per language and topic the earliest available article differs. A global constant would be misleading for most queries, so instead of publishing one figure APITube lets you measure the earliest date for your specific filters (sort ascending, per_page=1). That number is the one that actually applies to your use case.

Can I reach older articles on the Free plan?#

The date filter itself is not restricted on the Free plan — you can query any past range. What is limited is how deep you can paginate: Free stops after the first 5 pages (ER0173), which with per_page=250 is up to 1,250 articles per query. To page further into a historical window, upgrade to a paid plan. The earliest-article probe above still works on Free, since it only needs the first page.