How many news sources does APITube cover?#

APITube aggregates news from more than 500,000 sources — the news websites and publishers it crawls worldwide. That is the headline number, but for most projects the useful question is narrower: are the specific publishers I care about covered? Because APITube deliberately does not expose a single “download every source” endpoint, you confirm coverage by probing for the sources you need rather than pulling one giant list.

This page gives you the coverage figure, explains what counts as a source, and shows the three practical ways to check whether a given publisher — or a whole set of them — is in the pool.

How big is the source pool?#

The pool is 500,000+ news websites. A “source” in APITube is one publisher domain that the pipeline discovers and crawls — a site like reuters.com or bbc.co.uk, not an individual article. Each source carries its own metadata: a name, a domain, a type, a political bias label, and an OPR authority rank. Those sources publish across many languages and countries — for the language and country breakdown, see which languages and countries are supported.

Is there an endpoint that lists every source?#

No. There is no directory endpoint that dumps the full source list — a plain /v1/sources route is intentionally not available. With half a million publishers, a list-all call would be unusable anyway. Instead, you look sources up by prefix with the suggest endpoint, filter your search to specific domains, or aggregate the sources inside a result set with facets. The three sections below cover each approach.

How do I check whether a specific source is covered?#

Use the source suggest endpoint, which matches your typed text against source names and domains and returns the matches:

curl "https://api.apitube.io/v1/suggest/sources?prefix=bloomberg&api_key=YOUR_API_KEY"

The prefix parameter is required — omit it and the request returns ER0346. Each match comes back with its id, name, domain, type, and bias, plus a ready-made link to search that source:

[
  { "id": 4412, "name": "Bloomberg", "domain": "bloomberg.com", "bias": "center",
    "links": { "self": "https://api.apitube.io/v1/news/everything?source.id=4412" } }
]

The endpoint returns up to 100 matches and searches only active sources, so an empty result is a strong signal the publisher is not currently covered.

How do I restrict a search to specific sources?#

Once you know a publisher is covered, filter any search to it with source.domain. You can pass up to three domains as a comma-separated list:

curl "https://api.apitube.io/v1/news/everything?source.domain=bbc.co.uk,reuters.com&api_key=YOUR_API_KEY"

If a domain you pass is not in the pool, the request fails fast with 400 and error ER0214 (source domain name '…' not found) — a second way to test coverage. To go the other direction and exclude sources, use ignore.source.domain. The full filter is documented in how to filter news by source or domain.

How do I see which sources appear in my results?#

To measure the source diversity of a query — how many different publishers cover a topic and which ones dominate — add a facet on source.id. Set facet=true and name the field with facet.field:

curl "https://api.apitube.io/v1/news/everything?title=tesla&facet=true&facet.field=source.id&facet.limit=20&api_key=YOUR_API_KEY"

The response includes a facets block that counts articles per source for that query. facet.limit defaults to 10 and caps at 100, and facet.field is required when facet=true (otherwise ER0310). This turns “how many sources cover this?” into a number you can read straight off a single request.

Common Questions#

• Can I download the full list of sources?
• What counts as one source?
• What if the source I need is not covered?
• How current is the 500,000+ figure?

Can I download the full list of sources?#

No. APITube does not offer a list-all sources endpoint, so there is no single call that returns all 500,000+ publishers. The supported pattern is to look up sources you need by prefix with /v1/suggest/sources, then filter searches to them with source.domain. If you want to profile the sources behind a topic instead of a fixed list, facet a search on source.id and read the per-source counts.

What counts as one source?#

One source is one publisher domain that APITube crawls — a website such as reuters.com, counted once regardless of how many articles it publishes. Each source has a stable id, a domain, a name, a resource type, a political bias label, and an OPR authority rank, all of which you can see in a suggest result or use as search filters.

What if the source I need is not covered?#

If a suggest lookup returns nothing and source.domain rejects the domain with ER0214, the publisher is not yet in the pool — but you can ask for it to be added. See how to request a new news source. New sources are added over time, so a domain that is missing today may be covered later.

How current is the 500,000+ figure?#

500,000+ is the coverage figure APITube publishes, and the pool grows as new publishers are discovered, so treat it as a floor rather than a fixed cap. Because coverage changes, the reliable check for any single project is always to probe the exact publishers you need with /v1/suggest/sources and source.domain rather than rely on a headline count.