A video search API returns video results for a query across platforms — title, link, hosting platform, channel where shown, publish date and duration — from Google's video vertical rather than from any single site's own search.
That cross-platform property is the whole point. YouTube's API tells you about YouTube. This tells you what a searcher looking for a topic actually finds, which routinely includes Vimeo, Dailymotion, TikTok, news sites and embedded video on ordinary pages. If your question is "who owns video search for this term", one platform's API structurally cannot answer it.
$0.0004 per delivered video, up to 100 videos per run. Nothing delivered means nothing charged.
The query runs against Google's Videos vertical from an exit in your chosen country, and each result becomes a row. HTTP tier only — no browser is started, which keeps the run at $0.0004 per video.
The source field carries the hosting platform, and it is the field that makes this collector different from a YouTube integration: grouping a result set by source immediately shows you whether a topic is a YouTube monopoly or a genuinely contested surface. duration and date come through where the platform publishes them, which is most of the time for the major hosts and less often for embedded video on general sites.
query is required; country and lang set the market and the exit. Video results are strongly localised — the same query in the US and in Germany returns substantially different creators — so pin both for anything comparative. max_results caps delivery and cost.
One row per video. source is the hosting platform and is the field most worth aggregating on. channel is present when the platform exposes it in the result, which is reliable for YouTube and patchier elsewhere — nullable rather than guessed. date and duration are as published; a missing duration usually means the result is an embedded player on a general page rather than a native video listing.
$0.0004 per delivered video ($0.40 per 1,000). Nothing delivered means nothing charged, and the $2 monthly free credit covers roughly 5,000 videos here. Volume tiers take up to 30% off.
$0.0004 per delivered video, 40 cents per 1,000. Tracking 500 topic queries daily at 20 videos each is 10,000 rows, $4 a day at list price and less on a volume tier.
You pay per delivered video rather than per search, so narrow queries that return six results cost six units. Blocked pages and retries are never billed.
The YouTube Data API is a good API with two properties that decide whether it fits: it is YouTube-only, and it is quota-bound. Every project gets a daily quota measured in units, and search is one of the most expensive operations against it — which is why teams doing continuous topic monitoring hit the ceiling quickly and end up applying for quota increases rather than collecting data.
The deeper issue is scope. If you are analysing a channel you own, or pulling metadata for known video ids, the official API is the right tool and you should use it. If you are asking which videos rank for a topic across the open web, YouTube's index is one input among several and its internal ranking is not search ranking.
For YouTube-specific work at volume without quota mathematics, the YouTube scraper API and YouTube channel collectors cover that ground directly.
Doing this properly by hand means integrating every platform separately and then reconciling them, which is where the effort actually goes:
Reading the video vertical sidesteps all four: one request, one schema, and an order that reflects a real search result rather than a reconstruction.
Track which videos and which platforms rank for your topics, and whether your own content is present at all.
Watch what competitors publish and how quickly it surfaces in search, using date as the recency signal.
Group by source over time to see a topic migrating between YouTube, TikTok and news video.
Catch video coverage that mentions a brand or product on platforms nobody thought to monitor.
Up to 100 videos per run with pagination handled internally. Broader coverage is more queries, parallel up to your plan's rate limit — 60 requests/minute on pay-as-you-go, up to 1,200 on the top tier.
Legally: search results are public and collecting public data is generally lawful in most jurisdictions, but Google's terms restrict automated access and the analysis varies by country. Video content itself is copyrighted by its creators — this returns metadata and links, not media. Channel names can be personal data when the creator is an individual. Not legal advice.
The YouTube Data API is free within a daily quota, but it is YouTube-only and search burns quota fast. Here there is a free allowance of $2 a month with no card, about 5,000 delivered videos at $0.0004 each, across every platform Google indexes rather than one.
Scope. The YouTube Data API covers YouTube and ranks by YouTube's rules. This returns Google's video vertical, so results span YouTube, Vimeo, Dailymotion, TikTok, news sites and embedded players, ordered as a searcher actually saw them.
$0.0004 per delivered video — 40 cents per 1,000 — with no subscription and no quota to apply for. Volume tiers take up to 30% off; blocked pages and empty results are never billed.
No. It returns metadata and links: title, platform, channel, date and duration. Downloading media or pulling transcripts is a separate job with its own copyright considerations.
No. Your QuantumProxies key is the only credential. There is no Google Cloud project, no OAuth flow and no quota request.
Because not every platform exposes a channel in the search result. It is nullable rather than guessed — YouTube results almost always carry it, embedded video on general sites often does not.
There is no platform filter input, but source is on every row, so filtering after delivery is trivial. If you only ever want YouTube, the dedicated YouTube collectors are cheaper for that job because they do not pay for results you discard.
Skip the marketing. QuantumProxies.io publishes a machine-readable site map, Markdown for every page, a free MCP server, and APIs billed from the same balance as your proxies.
Connect in one command: npx -y quantumproxies-mcp