Romanian news API

What the coverage looks like

A Romance language in a region where most neighbours are Slavic, which shows in its coverage: extensive reporting on Central and Eastern European politics framed differently from either the Western or Slavic press. Outlet count is high relative to volume. Publication is noticeably bursty. Coverage clusters around announcements, parliamentary sessions and EU funding decisions, then falls away almost entirely between them, which means a seven-day window can return either a flood or almost nothing depending on where it lands. Anyone monitoring this language on a fixed short window will conclude the coverage is unreliable when what is actually unreliable is the sampling.

Outlet spread is the number worth watching here, not volume. 119 distinct publishers at roughly 15 articles each describes a very different information environment than the same total produced by a handful of prolific sources — and it is the difference between reaching independent reporting and reading one newsroom repeatedly.

In this sample the most active publisher is news.yam.md, carrying about 22% of Romanian articles on its own, and the three busiest — news.yam.md, stiripesurse.ro, mediafax.ro — account for roughly 28% between them. That is a comparatively flat distribution, which means volume here translates fairly directly into independent reporting rather than one newsroom publishing repeatedly.

What is the ISO 639-3 code for Romanian?

ron. Pass it as the language filter on any endpoint. Romanian is written in Latin with diacritics.

What does Latin with diacritics script mean for Romanian search results?

Inflected, with a definite article that attaches to the end of the noun rather than standing before it — so the "same" word appears with and without a suffix that a matcher will treat as unrelated. Diacritics are inconsistently applied and often absent entirely in headlines.

Which retrieval mode should I use for Romanian news?

Semantic mode, and do not assume diacritic-correct spelling in any query you construct. Widen time windows for niche subjects; the stream is mid-sized and thins quickly outside major topics.

Bound every request with a time range regardless of mode. Meaning-based retrieval always returns its best guesses, so an unbounded query produces a confident-looking result set anchored to nothing in particular. If results feed a model rather than a person, collapse to stories as well — one event covered by thirty outlets should cost you one item, not thirty.

Recent headlines in Română

Pulled from the index on 2026-07-31, shown in their original script so you can see what the API actually returns rather than a translated approximation of it.

• Detalii despre starea antrenorului de la Dinamo rănit grav în accidentul din Argeș: "Cred că vor să trimită elicopterul, se fac niște intervenții" ziare.com
• Accidente rutiere de microbuz, autobuz și autocar (cronologie selectivă 2024-2026) agerpres.ro
• În plin sezon al incendiilor, aproape 18% dintre fumătorii din Franța recunosc că aruncă chiștoace din mașină mediafax.ro
• Bumbești-Jiu: Și-a bătut soția, iar vecina a alertat polițiștii - Ştiri locale de ultima ora, stiri video gorjeanul.ro
• Motorina a trecut de 10,5 lei pe litru - Ştiri locale de ultima ora, stiri video gorjeanul.ro
• Amalgamarea voluntară câștigă încă o lună? Maia Sandu: „Oamenii se pot așeza și găsi soluții" news.yam.md

Outlets publishing in Romanian

The most active publishers in this language, by article count. Use these as an outlet filter when you want to pin a particular market or editorial position rather than taking the whole language.

news.yam.md · stiripesurse.ro · mediafax.ro · ziare.com · agerpres.ro · business24.ro · romaniatv.net · capital.ro · digi24.ro · adevarul.ro · bursa.ro · realitatea.net

What catches people out when querying Romanian?

The definite article attaching to the end of the noun is the specific trap. It means the same word appears in two forms that a string matcher treats as unrelated, and because the suffixed form is extremely common in headlines, a query built on the bare form misses much of the coverage. Missing diacritics compound the problem.

A worked case

Use semantic mode and never assume diacritic-correct text in anything you match against. For EU funding and infrastructure stories, widen the window to several weeks — these develop slowly and are reported in bursts around announcements rather than continuously.

Where this language leads

Good coverage of regional politics, EU funding and infrastructure, and Black Sea affairs. Business coverage is concentrated in a handful of outlets.

Who queries it, and why

Queried for EU funds, infrastructure and Black Sea affairs. A mid-sized stream that repays patience: the reporting is there but arrives in bursts around announcements rather than continuously, so short windows return misleadingly little.

Reading these numbers

Provisional. Measured from one ingestion adapter's public feed rather than from the index itself. Outlets and headlines are real; the counts are the right order of magnitude and are not coverage guarantees.